Brent crude touched $115.84 on Friday, its highest level since May. Houthi attacks on Saudi Arabia and fears of new US strikes on Iran drove the move.
Other energy prices were mixed, Trading Economics data show:
West Texas Intermediate (WTI) crude rose 1.35% on the day to $100.81. Heating oil gained 1.48% to $5.01 per gallon. US natural gas fell 2.86% to $2.72 per million British thermal units (MMBtu). Brent Crude Daily Price Chart Showing the Climb to its Highest Level Since May. Source: TradingViewThe rally landed on markets still shaken by a tech selloff. Crypto also slipped as investors pulled back from risk.
Gulf Attacks Put a Risk Premium Back on OilYemen’s Houthis said they hit Saudi energy targets, including an Aramco site in Yanbu and the Shaybah oilfield. Aramco later ordered staff out of its Dhahran headquarters on the Gulf coast.
OilPrice noted that this suggests the group can now strike well beyond its usual Red Sea targets. A pipeline shutdown in Libya also cut output at two fields.
ING analysts cited reports that Trump is weighing fresh Iran strikes as a second US carrier deploys. The bank now calls its earlier $80 fourth-quarter Brent forecast unattainable. Brent first reached a five-month high earlier this week, BeInCrypto reported.
“If anything, re-escalation seems the most likely path. As a result, our expectation that ICE Brent would average $80/bbl over 4Q26 is looking unattainable. Even if tensions don’t re-escalate, the mere possibility is likely to keep a sizeable geopolitical risk premium embedded in prices,” the analysts said.
Still, commodity trader Jack Prandelli sees a split. He noted Dated Brent, the physical cargo price, slid from about $135 in early October to around $118.
The oil market is sending 2 different messages.
Dated Brent, the price of physical cargoes, has dropped from about $135 in early October to around $118.
Brent futures have gone the other way, up to about $112.
The gap between them has shrunk from roughly $30 to under $10 in 3… pic.twitter.com/BXInybZC0i
— Jack Prandelli (@jackprandelli) October 22, 2026 Follow us on X to get the latest news as it happens
Tech Selling and Rate Fears Weigh on Risk AssetsThe oil rally arrived amid a sharp drop on Wall Street. On Thursday, the Nasdaq Composite fell 2.6% to 26,008.86, while the S&P 500 lost 1.21%.
Meta led the decline, falling nearly 7% after raising its 2026 spending floor to $135 billion.
Crypto did not escape Thursday’s selling on Wall Street. That spillover shows in BeInCrypto Markets data. Total crypto market capitalization fell 3.74% in 24 hours to $2.76 trillion. Ethereum (ETH), XRP (XRP), and TRON (TRX) were among the major assets trading lower.
Analyst Benjamin Cowen tied that volatility back to oil.
Oil adds pressure through interest rates. Higher crude feeds inflation expectations and strengthens the case for tighter Fed policy.
Traders still saw about an 83% chance that the Fed holds rates at its October 27-28 meeting. Renewed energy price gains, however, could shift those odds.
Next week brings two tests. The Fed announces its rate decision on October 28, and Microsoft, Amazon, and Alphabet report earnings.
Whether Brent holds above $115 into those events may decide how quickly risk appetite returns.
Subscribe to our YouTube channel to watch leaders and journalists provide expert insights
Privacy Policy . Terms and Conditions . MiCA Compliance Whitepaper
The $NAORIS token and the Naoris / Naoris Protocol name and logo are used under license. The underlying network technology, security software and related services are provided by Naoris Quantum Protocol Inc. under license. For any software or service enquiries, please contact [email protected].
Bitcoin, Ethereum, XRP Gain, but Dogecoin Slips Amid Spike in Rate Hike Odds: Analyst Says Everything Is ‘Fine’ for BTC if It Trades Above This Key Level
Altın fiyatları, küresel piyasalarda güçlenen ABD doları, yükselen petrol fiyatları ve tahvil getirilerindeki hareketlilik nedeniyle dalgalı bir seyir izliyor. Ons altın 4.190 dolar seviyelerinde işlem görürken yatırımcıların odağı ABD’den gelecek kritik enflasyon verilerine çevrildi. Uzmanlar, kısa vadede altın fiyatlarında yeni bir düşüş yaşanabileceği konusunda uyarırken özellikle 4.000 dolar seviyesinin önemine dikkat çekiyor. Piyasalarda Fed’in faiz politikasına ilişkin beklentiler de altının yönünü belirleyebilecek önemli faktörler arasında yer alıyor.
Altın Fiyatlarında Son Durum Spot altın, saat 07.27 itibarıyla yüzde 0,50 yükselişle ons başına yaklaşık 4.190 dolardan işlem görerek dalgalı seyrini sürdürdü. İç piyasada ise dolar/TL kurundaki hareketlilik ve küresel altın fiyatlarının etkisiyle gram altın 6.595 TL alış ve 6.651 TL satış seviyelerinde fiyatlandı. Güçlenen ABD doları ve yüksek tahvil getirileri, yatırımcıların faiz getirisi sağlamayan altına yönelik talebini sınırlarken petrol fiyatlarındaki yükseliş enflasyon endişelerini artırıyor. Uzmanlar, küresel ekonomik belirsizliklerin ve Fed’in faiz politikasına ilişkin beklentilerin önümüzdeki dönemde altın fiyatları üzerinde belirleyici olabileceğine dikkat çekiyor.
İlginizi Çekebilir: Zcash’ten Kuantum Tehdidine Karşı Kritik Hamle!
Yatırımcılar, ABD’de açıklanacak eylül ayı tüketici enflasyonu verilerine odaklandı. Ardından gelecek üretici enflasyonu ve perakende satış rakamlarının da Fed’in para politikasına yönelik beklentileri şekillendirmesi bekleniyor. Piyasalarda Fed’in ekim ayında faiz artırma olasılığı yaklaşık yüzde 17, yıl sonuna kadar en az bir faiz artırımı gerçekleştirme ihtimali ise yüzde 87 olarak fiyatlanıyor. Beklentilerin üzerinde açıklanabilecek enflasyon verileri, faizlerin daha uzun süre yüksek kalacağı düşüncesini güçlendirerek altın fiyatlarında satış baskısını artırabilir.
Uzmanlardan Altın İçin 4.000 Dolar Uyarısı Kitco News’in aktardığına göre Trade Nation Kıdemli Piyasa Analisti David Morrison, altının güçlü bir yükseliş ivmesi kazanabilmesi için 4.000 dolar destek seviyesinin yeniden test edilmesi gerekebileceğini belirtti. UBS analistleri de kısa vadede aynı seviyeye doğru geri çekilme riskine dikkat çekerken, Eylül 2027’ye kadar ons altının 5.400 dolara ulaşabileceğini öngörüyor. Bu değerlendirmeler, kısa vadeli düşüş riskine rağmen bazı kurumların altının uzun vadeli görünümüne yönelik olumlu beklentilerini koruduğunu gösteriyor.
FOREX.com piyasa analisti Fawad Razaqzada, son değerlendirmesinde doların güçlü kalması ve faiz oranlarının yüksek seyretmesi nedeniyle altının toparlanmasının “biraz ivme kaybettiğini” vurguladı.
Petrol fiyatlarındaki oynaklık da enflasyon endişelerini canlı tutarken yatırımcıların Fed’in gelecek faiz kararlarına yönelik beklentileri yakından takip etmesi bekleniyor.
Değerlendirme Altın fiyatları, ABD enflasyon verileri öncesinde kritik bir döneme girerken uzmanlar 4.000 dolar seviyesini önemli bir destek noktası olarak değerlendiriyor. Güçlü dolar ve yüksek faiz beklentileri kısa vadede düşüş riskini artırırken, merkez bankalarının alımları ve yatırım talebi uzun vadeli görünümü desteklemeye devam ediyor. Altının sonraki yönünde ABD ekonomik verileri ve Fed’in para politikası belirleyici olacak.
Son Dakika kripto para haberleri için hemen tıkla.
Konu ile ilgili yorumlarınızı bize yazabilirsiniz. Ayrıca, bu tarz bilgilendirici içeriklerin devamının gelmesini isterseniz, bizleri Telegram, Youtube ve Twitter kanallarımızdan takip edebilirsiniz.
Billionaire investor Ray Dalio is warning that the US Treasury market is exposed to a drop in demand from China and Japan.
The Bridgewater Associates founder says the US depends on foreign capital for about a third of its debt, with much of that money coming from Japan and China, reports Bloomberg.
Dalio tells Bloomberg Television that geopolitical tensions mean China no longer wants to keep adding US debt to its holdings.
He says Japan has lent “a lot of money” that it now wants to bring back home.
Since the start of the year, China and Japan have reduced their holdings of US treasuries by a combined $198.7 billion. Japan, which is the largest foreign creditor to the US, held $1.1039 trillion in US treasuries as of July. China is the third-largest foreign creditor to the US and as of July held $618 billion in US treasuries.
Dalio also repeated his warning that the US could face a debt crisis within three years.
His comments come as a global bond selloff pushes government borrowing costs to multi-year highs.
The yield on the 10-year US Treasury is hovering around 5.3%, a level not seen since 2002.
Inflation, fiscal spending and resilient economic growth have weakened demand for bonds around the world.
Bond prices have kept falling this week, extending a slide that has lasted for months.
Crypto whale sets 10 key targets first: Amid multiple macro headwinds, BTC has only corrected around 5% and will continue holding as long as it does not drop below $79,000.
Whale "Set 10 Big Goals" (@jasonleo) posted late at night, noting that the most notable aspect of Bitcoin’s (BTC) current correction is not the magnitude of the drop, but rather "why it only fell this much". He pointed out that recent macro headwinds include the 30-year U.S. Treasury yield approaching 5.7%, the 10-year yield around 5.3%, rising market expectations for another interest rate hike this year, and resurfacing oil price and inflation pressures. Despite this confluence of negative factors, BTC has only corrected by roughly 5% so far. Jasonleo argues that an asset’s price reaction to news is meaningful in itself. In the past, a single macro headwind could trigger a BTC drop of over 10%; this time, however, under the combined pressure of multiple headwinds, the price has not suffered significant damage—an indication that the current uptrend has notable resilience. Comparing BTC’s current trajectory to its prior bottom near $58,000, he says both periods share the trait of "selling pressure easing after all negative factors are fully priced in", while the market remains in a strong trend. He added that he does not expect BTC to plummet directly to $78,000 or even $74,000, setting $79,000 as the watch level for position adjustments: if BTC breaks below $79,000, he will start reducing positions; if it closes below $78,000 on a daily basis, he will exit all remaining long positions to wait for the next entry opportunity; if the key level holds, he will continue holding long positions.
3 hours ago
Papertrade opens pre-deposits, official trading expected to launch on October 10.
Synthetic perpetual protocol Papertrade, built on HyperEVM, officially announced its pre-deposit feature is now live. Users can complete deposits and account creation within the next two days. The official noted that account creation and deposits will be significantly restricted once trading launches; depositing early helps avoid opening congestion, though there are no additional benefits for depositing during the pre-deposit period. Papertrade is a fully on-chain synthetic perpetual trading platform deployed on HyperEVM. It fetches prices from the Best Bid/Offer (BBO) mid-price of Hyperliquid’s order book, allowing users to trade directly against the protocol’s liquidity providers (LPs) instead of relying on traditional order book matching. The protocol’s key features include high leverage, zero slippage, and no funding fees. Its native token PAPER has an initial supply of zero, with no pre-mining, team, or venture capital (VC) allocations. The only source of token minting is user losses or liquidations. During the initial launch phase, the token cannot be freely transferred and is primarily used for staking. Per the official timeline, full trading will begin approximately one hour after the expected HyperEVM network upgrade on October 10. The exact time will be announced via Hyperliquid’s Discord. During the opening phase, trading must be conducted through the official website frontend and whitelist relays; direct contract interactions will be opened once congestion eases. For more related reading, see: "Hyperliquid Is Getting Its Own Layer 2: Key Details You Need to Know".
3 hours ago
The US stock storage sector’s decline has widened further, with Sandisk and SK Hynix ADRs falling more than 5%.
According to market data from BIT (bit.com), losses in the U.S. stock market storage sector have widened further, with individual stocks recording the following declines: Seagate Technology (STX) fell 3.94%; Western Digital (WDC) dropped 3.05%; SanDisk (SNDK) declined 5.4%; Micron Technology (MU) fell 4.21%; and SK Hynix ADR (SKHY) decreased 5.45%.
3 hours ago
The Nasdaq's decline widened to 1%, while the Philadelphia Semiconductor Index fell more than 2%.
According to BIT (bit.com) market data, the Nasdaq’s decline widened to 1%, while the Philadelphia Semiconductor Index fell more than 2%. Intel and Marvell dropped over 4%, AMD and Micron declined more than 3%, and Nvidia fell nearly 2%. Earlier news: OpenAI’s annual revenue is $20 billion lower than previously reported, reaching $50 billion as of the end of September, compared to the widely cited $70 billion figure.
3 hours ago
OpenAI’s annual revenue is $20 billion lower than previously reported, with the latest figure standing at $50 billion.
Beating AI Express, according to the Financial Times, OpenAI recently disclosed to investors that its annualized revenue stood at nearly $50 billion as of the end of September, significantly lower than the $70 billion reported at the end of last month. Insiders said the discrepancy arises because investors tried to directly compare OpenAI’s annualized revenue with that of its rival Anthropic, as the two companies use different accounting standards: Anthropic includes sales revenue generated via cloud partners such as AWS and Google Cloud in its revenue, while OpenAI does not. Earlier, investors adjusted OpenAI’s revenue standard to arrive at a $40 billion figure for its August annualized revenue, then combined this with the company’s disclosed revenue growth of over 70% to investors to calculate the $70 billion number. OpenAI declined to comment.
3 hours ago
Federal Reserve Chair Walsh is scheduled to deliver a speech at the IMF Annual Meeting in Bangkok on October 16.
According to the official agenda for the International Monetary Fund (IMF) Annual Meeting in Bangkok, Federal Reserve Chair Walsh has confirmed his attendance at the event and is scheduled to deliver a public speech on October 16 local time. His trip has been formally included in the meeting’s agenda by the IMF Bangkok Annual Meeting Organizing Committee. As the head of a leading global monetary policy-making institution, Walsh’s appearance has drawn significant attention. During the conference, his speech will focus on elaborating on the current macroeconomic situation and monetary policy framework. The IMF has now released the relevant schedule in accordance with established procedures. All speaking arrangements for participating guests have been posted on the official platform, and specific details of the speeches will be subject to on-site disclosures.
.frax would connect a globally recognized internet domain with onchain ownership, payments, and AI-native commerce
, /PRNewswire/ -- Frax today became the first crypto company to have its own top-level domain application revealed by ICANN, with .frax officially published as part of the 2026 New Generic Top-Level Domains Program. If successfully delegated, .frax would become a globally recognized internet top-level domain, similar to .com or .org.
Frax is the first crypto company to have its own top-level domain application revealed by ICANN. Frax plans to make .frax the first end-to-end top-level domain designed to bridge the traditional internet and the onchain economy. Users would be able to own and manage .frax domains through Frax Name Service (FNS) on Frax's blockchain Fraxtal while those same domains resolve through the global Domain Name System. FNS registrations also feed the Frax Burn Engine, permanently burning FRAX and allowing the ecosystem's native digital commodity to capture value as domain usage grows.
"Crypto naming systems made internet identity ownable. ICANN domains made it universally reachable. .frax brings those two worlds together," said Sam Kazemian, Founder of Frax. "Our goal is to build the first internet namespace where ownership lives onchain but the domain works everywhere. From there, domains can evolve into programmable financial endpoints for people, businesses, applications, and AI agents."
The initiative also extends the financial network Frax is building around frxUSD and FraxNet. FraxNet connects financial institutions, fintechs, and companies through a shared network for digital money and payments, with each new participant expanding the reach and utility of frxUSD.
Frax also plans to explore agentic and machine-payment functionality for .frax domains, including support for emerging standards such as x402. This could enable websites, applications, and autonomous software agents to identify counterparties, access services, and initiate stablecoin payments programmatically.
The .frax application was submitted with support from MarkMonitor, an ICANN-accredited corporate domain registrar.
About Frax
Frax is a U.S.-based financial technology company building the network for secure, global, and programmable money. At its core is frxUSD, a fully backed digital dollar, and FraxNet, a financial platform connecting institutions, fintechs, and enterprises through a shared network for digital money, payments, lending, credit, and yields. Frax provides stablecoin and crypto payments infrastructure to leading financial institutions, companies, and fintechs. Since 2020, Frax has powered billions of dollars in onchain capital and activity across more than five years of serving users. Frax was ranked #5 in Stablecoins in the 2026 Fortune Crypto 100. Learn more at Frax.com.
Media Contact for Frax
Gary Bird
FortyThree, Inc.
831.888.9011
[email protected]
Bitcoin fell through $84,000 in two legs, the first in the Asia session and the second at the U.S. open, and has not reclaimed it. A Bitfinex research note circulated hours earlier had called that level the market's largest cost-basis cluster. Funding rates sat near zero across OKX, Deribit and Hyperliquid through the decline.
Bitcoin fell through $84,000 in two legs, the first in the Asia session and the second at the U.S. open, and has not reclaimed it. A Bitfinex research note circulated hours earlier had called that level the market's largest cost-basis cluster. Funding rates sat near zero across OKX, Deribit and Hyperliquid through the decline.
Loading this article…
Free newsletters
Good reporting. Delivered.Choose your coverage: crypto and DeFi, the future of finance, or both.
The Nigeria CERT advisory codifies Tenet Security's DEF CON 34 research on MCP integration vectors as a formal national threat. Three new attack surfaces (Cloudflare, Datadog, Sentry) and an 85% success rate against Claude Code, Cursor, and Codex force the trust-through-defaults conversation into the national CERT domain.
The Nigeria Computer Emergency Response Team (ngCERT) issued an advisory on October 6, 2026, formalizing Ghostjacking as a government-level threat. This marks a shift from academic curiosity to institutional recognition of the risks inherent in autonomous systems. The advisory codifies the threat landscape previously detailed by Tenet Security researchers Barak Sternberg, Nevo Poran, and Ron Bobrov during their DEF CON 34 presentation on August 9, 2026.
Ghostjacking exploits the Model Context Protocol (MCP) and its reliance on implicit trust. The attack vector functions because AI agents are designed to treat external data as instructions. When integrated with platforms like Cloudflare, Datadog, or Sentry, agents often process retrieved data as trusted input. Because the resulting actions—such as code execution or cloud credential access—fall within the agent’s authorized scope, conventional security controls like EDR, WAF, IAM, and VPNs remain silent. The agent is simply executing its programmed instructions, even when those instructions originate from a malicious injection.
Researchers have identified three primary MCP integration vectors:
Cloudflare: Facilitates domain hijacking. Datadog: Enables code execution combined with cloud credential theft. Sentry: Functions as an insider-style agent vouching mechanism. The scale of the vulnerability is quantifiable. CSA Labs research from June 2026 demonstrated an 85% exploitation success rate across Claude Code, Cursor, and Codex by leveraging Sentry DSN injection. By the time of the DEF CON 34 presentation in August, researchers observed a 90% success rate against Claude Code specifically when using Cloudflare’s recommended setup. Furthermore, an audit identified 2,388 organizations with injectable, publicly exposed Sentry DSNs, including 71 sites within the Tranco top-1M list. According to Tenet Security researchers, Sentry was notified of these findings on June 3, 2026, but declined to implement root-cause remediation, characterizing the issue as technically not defensible at the platform level.
Beyond these MCP integration vectors, researchers identified a separate vulnerability in the Claude Desktop egress sandbox. This zero-day involved JWT cross-reuse in the envoy egress gateway, allowing for session token reuse across containers. Anthropic patched this issue prior to the DEF CON 34 presentation, and no CVE was issued. This incident highlights that while MCP-based attacks are the primary focus of the Ghostjacking advisory, the underlying infrastructure of desktop-based AI agents remains a complex surface for potential privilege escalation.
For organizations deploying AI agents, the immediate challenge is the trust-through-defaults pattern inherent in monitoring and integration layers. Because detection tools are largely ineffective against actions that appear legitimate to the system, prevention must occur at the configuration level. Tenet Security has released agent-jackstop, an open-source tool providing hardening configurations for Cursor and Claude Code. It is critical to note that this is not a detection tool; it is a preventative measure intended to reduce the blast radius of a successful injection by restricting what an agent can do by default.
This advisory serves as a follow-up to our initial coverage of Agentjacking. The formalization of Ghostjacking by ngCERT underscores the transition toward a structured understanding of how AI agents can be subverted without triggering traditional security alerts. This development aligns with our previous reporting on ClawSecure MCP risks, the complexities of MCP OAuth implementations, and the broader Agent Identity Layer risks. The reliance on authorized actions means that until the industry moves away from implicit trust in MCP integrations, the burden of security rests on granular, restrictive configuration rather than reactive monitoring.
Ethoswarm Heath Callahan works for Forkast.
Minds can also work for you.
Minds are persistent AI beings with instincts, identity, and a job.
Awaken one on Ethoswarm.
The market structure is significantly different today, said one of the analysts we talked to.
Bitcoin enters the final quarter of the year after a powerful recovery in the third quarter, but analysts warn against expecting another straight-line rally. Instead, they pointed to some key factors that could impact BTC and the overall market in the following three months.
Some of them include the Federal Reserve, Treasury-market liquidity, ETF flows, geopolitics, and BTC’s ability to clear $87,500, which remains its most significant obstacle on the path forward.
Q3 May Be Hard to Repeat Although Q3 began with another leg down to under $58,000, which became BTC’s lowest price tag in a year and a half, the subsequent three months were a lot more positive. The cryptocurrency rebounded immediately and broke out in mid-August to over $80,000. Its rise continued and managed to close the quarter with a massive 43% surge.
Iliya Kalchev, Nexo Dispatch Analyst, described the three-month period as both a recovery phase and a breakout milestone. He argued that the most important catalyst arrived from the bond market after the US Treasury increased the size of its long-end bond buyback operations in August.
The asset indeed jumped by 7% on August 19 and rocketed by over 20% in the following several days. Spot Bitcoin ETF flows immediately turned positive and even flipped into the green on a year-to-date basis. Meanwhile, relatively subdued perpetual funding suggested the rally was driven more by spot demand than excessive leverage, Kalchev added.
Nevertheless, the analyst cautioned against assuming Q4 will simply extend Q3’s pace. Although the cryptocurrency has finished Q4 higher in nine of the past 15 years, the median gains are only around 9%, while the much larger average has been distorted by spectacular years such as 2013 and 2017.
Alex Kozenko, CMO at WhiteBIT, issued a similar warning:
You may also like: Is Bitcoin’s $85K Consolidation the Calm Before the Storm Amid Rising Middle East Tensions? 4 Cryptocurrencies to Watch This Week: BTC, HYPE, ZEC, and KAS Face Crucial Tests Citi Turns More Bullish on Bitcoin and Strategy: Here Are the New Targets “Today, the market structure is different: institutional participation has become more prominent, and flows through regulated investment products have become yet another source of influence on market dynamics. Over the next three months, I would primarily focus on liquidity, institutional activity, and the overall macroeconomic environment.”
Fed and $87.5K Although the overall market situation changed slightly after the weaker-than-expected US jobs report from Friday, Lacie Zhang, Research Analyst at Bitget Wallet, told CryptoPotato that she still believes the Fed will hike rates again by 25 basis points on October 28. This would put the target range at 4.00%-4.25% after the September increase, which was the first in over three years.
Kalchev also highlighted the Fed as the biggest Q4 variable, although the latest softer core PCE reading, alongside the aforementioned jobs report, reduced some of the immediate pressure for additional tightening. Geopolitical developments, though, could complicate the picture further, especially if energy prices keep feeding inflation.
According to Zhang, $87,500 remains the most crucial obstacle in BTC’s path to a broader recovery. A break above it could increase the likelihood of a short squeeze. In contrast, she identified the $82,000-$82,500 support range as the key downside zone, and losing it could accelerate a move below $80,000.
Apple tightens macOS Full Disk Access controls citing AI agent risks – the first time a major OS vendor has updated core permissions specifically because of agents.
On October 2, 2026, Apple announced it will tighten the controls around macOS Full Disk Access, explicitly citing the risks that autonomous AI agents pose to user data. This is the first time a major operating system vendor has updated its core access permissions specifically because of agents – not malware, not nation-state threats, but the architectural reality that software designed to act on your behalf needs different boundaries than software designed to back up your hard drive.
The problem starts with what Full Disk Access actually does. Originally carved out so backup utilities like SuperDuper and Carbon Copy Cloner could read the entire disk, FDA bypasses Apple’s per-app TCC privacy controls – the system that normally gates access to Camera, Microphone, Photos, and other sensitive resources on a per-app basis. When an application holds FDA, it can reach Mail, Messages, Safari history, contacts, photos, and Time Machine backups. For a backup tool, that is the job. For an always-on AI agent, it is the entire surface area of your digital life.
Apple’s Developer News post states the company will “introduce additional controls to ensure that users who genuinely wish to grant an app this extraordinary level of access can only do so with very explicit user action.” The language is careful – this is a consent change, not a technical ban on FDA – but the framing is unambiguous: “As AI agents become increasingly capable and autonomous, the risks associated with this level of access will grow substantially.”
The announcement follows two incidents that exposed how broadly agents can reach once they hold FDA. In September, Inc. columnist Jason Aten reported that Meta’s Muse AI agent had read his private Messages without him knowingly granting permission – a claim Meta disputed, stating that three explicit user actions are required (enabling FDA, enabling the in-app Messages connector, and responding to a macOS system dialog). Separately, Wired reported a flaw in the ChatGPT Mac app, discovered by the Objective-See Foundation and patched September 25, that could have let attackers access chat logs and browser sessions by exploiting a trusted script interpreter component.
Patrick Wardle, the Objective-See Foundation researcher who found the ChatGPT vulnerability, framed the structural risk plainly: “Agents need a lot of access to do their job. They are like the building manager who has access to the keys to all the rooms. So if they can be corrupted or subverted, that’s super problematic. It can mean that unprivileged code could then potentially have access to all the things.”
The affected agents are not edge cases. Meta’s Muse, OpenAI’s Dots, the ChatGPT Mac app, Claude for Mac, OpenClaw, and Hermes Agent all commonly request Full Disk Access to deliver their desktop capabilities. Apple’s move forces a question that the enterprise security layers Forkast has covered have not had to answer at this level: who is the final arbiter of what an application can see?
Prior coverage has tracked the security perimeter forming at the enterprise layer. The execution-layer gateway governs which APIs agents call and which data they touch. Kubernetes Agent Sandbox provides container-level isolation for agent workloads. Aembit’s XAA enforcement point brokers identity-based access control. Each layer is real, each is shipping, and each operates above the operating system. Apple’s intervention is below all of them – at the point where the OS itself decides what an application can reach.
There is no equivalent policy change on Windows or Linux. Windows relies on UAC and AppContainer for privilege elevation and per-app capability gating; desktop Win32 apps still run with the user’s full token by default. Linux offers AppArmor, SELinux, and Flatpak portals for sandboxing, but most desktop distributions do not enforce a unified permission broker for desktop applications. No major operating system yet ships a purpose-built permission model for autonomous AI agents. Apple is first, and it is doing it by retrofitting the consent flow around a legacy permission rather than building a new one from scratch.
Apple has not announced which macOS version will ship the new controls or when. The gap between announcing intent and shipping implementation is where developers will need to make decisions – about how much access their agents request, how they communicate that access to users, and whether the “all-or-nothing” model of FDA can survive an era where agents need to touch everything but should be trusted with only some of it.
Ethoswarm Blair Hayes works for Forkast.
Minds can also work for you.
Minds are persistent AI beings with instincts, identity, and a job.
Awaken one on Ethoswarm.
Delta Electronics and Nvidia are joining forces to build the brains, and the power supply, behind self-driving cars.
Delta announced a strategic collaboration with Nvidia on September 29, 2026, focused on developing next-generation Level 4 autonomous driving systems and robotaxis.
What the partnership actually covers The announcement came from Delta in Fremont, California. The company is publicly traded on the Taiwan Stock Exchange under the code 2308.
The deal combines two very different skill sets. Delta brings its experience in automotive energy management and electric vehicle powertrain systems. Nvidia brings Hyperion, its reference platform for autonomous vehicles.
Advertisement
On the technical side, Hyperion includes dual Nvidia DRIVE AGX Thor compute systems and a standardized sensor architecture.
The platform also includes Halos, Nvidia’s safety system. Halos is designed to support ISO 26262 ASIL-D compliance. ISO 26262 is the functional safety standard for road vehicles, and ASIL-D is its most demanding tier, reserved for systems where a failure could seriously hurt people.
The collaboration centers on two areas: power electronics and in-vehicle high-performance computing for Level 4 systems. Delta will also contribute capabilities in high-performance computing and system integration, working alongside Nvidia’s broader ecosystem.
What Level 4 means, and why power matters Level 4 sits near the top of the industry’s autonomy ladder. A Level 4 vehicle handles driving on its own within defined conditions, without needing a human ready to grab the wheel.
Executives from both companies stressed that pairing Delta’s multi-year EV experience with Nvidia’s scalable, safety-focused autonomy technology will benefit the automotive market. No direct financial terms accompanied that optimism.
Background: Delta’s road into electric mobility Delta committed to electric mobility several years ago, and the company points to that track record as the foundation for this deal.
For Nvidia, the collaboration lines up with a goal it has been pursuing for some time: production-ready Level 4 solutions. The topic has come up repeatedly in technical presentations at Nvidia’s GTC events, the company’s annual showcase for developers and partners.
Neither company has revealed specific commercial products, timelines or financial terms tied to the collaboration.
What this means for the autonomous driving race What to watch next: whether Delta and Nvidia name specific vehicle programs or automaker customers, whether joint hardware appears at a future GTC event, and whether Halos-backed systems move toward formal safety certification.
Disclosure: This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our Editorial Policy.
October now begins with one of the strongest seasonal records of the year. However, market breadth has dropped to its lowest level since May 2025.
S&P 500 Leaves Its Worst Month BehindBarchart data shows September is the only month with a negative average return for the SPDR S&P 500 ETF (SPY) since 2010. The fund has lost 0.48% in an average September.
October, in contrast, ranks as the third-best month. SPY has averaged a 2.27% gain, behind only November at 3.09% and July at 2.79%.
S&P 500 ETF (SPY) average monthly returns since 2010 / Source: BarchartSeptember’s record is more mixed than its reputation suggests. SPY rose in 10 of the last 17 Septembers, including a 0.26% gain this year. A few sharp drops, such as 9.24% in 2022 and 6.94% in 2011, pulled the average lower.
Historically, October has often rebounded after a weak September. In 2011, a 6.94% September loss gave way to a 10.91% October rally. Similar recoveries followed in 2021 and 2022.
Still, the pattern offers no guarantee. Three of the last six Octobers ended lower, in 2020, 2023, and 2024. SPY enters the month up 13.37% for the year.
75% of S&P 500 Stocks Fell in SeptemberThe index’s small gain masked weakness across most of its members. FactSet data, visualized by Deena Zaidi, shows about 75% of S&P 500 stocks closed September lower.
Losses reached every sector as Treasury yields climbed during the month. Banks formed one of the largest red blocks, with JPMorgan, Bank of America, and Wells Fargo all declining. Software names, including Salesforce, Adobe, and Oracle, also fell.
Meanwhile, the gains came from a few pockets. Micron, Apple, and Dell rose in technology, extending the market’s reliance on AI hardware. GE Vernova and Eli Lilly also posted gains.
As a result, a narrow group of large companies kept the index afloat while most stocks retreated.
S&P 500 Breadth Hits Its Weakest Level Since May 2025The 200-day moving average measures a stock’s long-term trend. Only 40.55% of S&P 500 stocks now trade above it, meaning nearly 60% sit below.
The reading fell 2.59 points on Sept. 30 alone. It peaked near 73% in August, so breadth has dropped about 32 points in roughly six weeks.
Share of S&P 500 stocks above their 200-day moving average / Source: BarchartThe indicator now sits below its April 2026 low of about 42%. Notably, that earlier trough came before a rally that lifted breadth to its August high.
What October Could BringSeasonality favors the S&P 500 in October. However, the index enters the month with its weakest participation in more than a year.
A broad recovery would likely require more stocks to reclaim their 200-day averages. Without that, the index may keep relying on a small group of large caps.
Disclaimer
This Analysis reflects BeInCrypto's editorial interpretation of information and data available at publication and may become outdated. It is general and non-personalised, does not consider your circumstances, and is not investment research, financial, investment, legal or tax advice, an offer or a recommendation. Forecasts, targets, technical analysis and forward-looking statements are uncertain and may not materialise; past performance and indicators do not predict future results. References to assets, products or providers do not imply endorsement. Crypto-assets and financial products may result in total loss. Verify material information before acting.
Google’s Gemini 4 Argon has matched OpenAI’s GPT-6 Astra on the Artificial Analysis Intelligence Index, scoring 53 points. At current launch prices, it also costs less to run per test task.
Google DeepMind announced the model on September 30. For now, only a group of trusted cyber defenders can use it.
Introducing Gemini 4 Argon – our new frontier model.
It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program. pic.twitter.com/X8acOWJOSF
— Google DeepMind (@GoogleDeepMind) September 30, 2026 Argon Puts Google Back Among the Top 3 LabsArtificial Analysis, an independent benchmarking firm, said the score puts Google back among the top 3 AI labs. Argon also sits 1 point above OpenAI’s GPT-6.1 Sol, which scored 52.
How Google Gemini Argon Performs. Source: X/Artificial AnalysisThe model scored 23 points higher than Gemini 3.1 Pro Preview, Google’s previous non-Flash model. Earlier in September, Google released Gemini 3.8 Flash alongside a cybersecurity variant. For comparison, SpaceXAI’s Grok 4.7 scored 46 on the same index in September.
Artificial Analysis credited Argon’s gains to fewer hallucinations and stronger agentic skills. On its AA-Omniscience test, Argon posted a 15% hallucination rate, compared with 51% for Astra. However, Argon answered 50% of those questions correctly, 13 points below Astra.
In agentic work, Argon led AutomationBench-AA at 77.5%, ahead of Anthropic’s Claude Sonnet 5.5 at 71.3%. Still, it trailed Sonnet 5.5, Claude Opus 5.5, and Astra on Terminal Bench 4.
Follow us on X to get the latest news as it happens
The Smaller Bill Runs on a Launch DiscountGoogle has set introductory pricing at $2 per million input tokens and $10 per million output tokens. Those rates double to $4 and $20 once the promotion ends.
At the discounted prices, Artificial Analysis measured Argon’s cost at $1.99 per Index task, against $3.26 for Astra. However, at standard rates, the figure rises to $3.98, about 1.2 times Astra’s cost.
The savings come from cheaper tokens, since Argon uses more of them. It averaged 62,000 output tokens per task, compared with 27,000 for Astra. The model also has a 1 million token output limit.
“To support Gemini 4 Argon’s capabilities across longer, more complex use cases, we are significantly expanding the model’s output token limit to an industry-leading 1M tokens, up from the previous 64K tokens,” the team said.
Cyber Defenders Get the First Look at Gemini 4 ArgonGoogle CEO Sundar Pichai said teams across Google already use it heavily, with work ranging from coding to quantum computing.
“Importantly, Argon has frontier safeguards, and we are rolling it out responsibly – it’s with the US gov’t and going to a set of trusted cyber defenders through our Fairwind Program today,” he noted.
Google’s launch post says Argon can find, validate, and patch serious software flaws on its own. For that reason, trusted defenders and Google’s internal teams will receive it without cyber guardrails.
Paid API customers and Google AI Ultra subscribers will get access first when the rollout widens. Google has not said whether the launch discount will still apply by then.
The firm also said it is engaged in the US government’s voluntary pre-release access process.
Subscribe to our YouTube channel to watch leaders and journalists provide expert insights
BloFin, a top-tier cryptocurrency exchange built for ambitious traders and serious capital, is returning as Title Sponsor of TOKEN2049 Singapore, marking a major milestone for the company as it celebrates its third anniversary and enters its NEXT WHALE ERA.
For BloFin, over the past three years, the crypto market has moved through a deep bear cycle, recovery, and multiple moments when the industry’s credibility was tested; the teams still building are the ones still here. That conviction sits at the center of BloFin’s 2026 direction.
While broader crypto sentiment remains muted, BloFin believes the market’s future is bigger than crypto alone. It is programmable money, open settlement, global markets that never close, and an industry increasingly converging with AI. As autonomous agents begin to hold, move, and trade value, and as traders rely more heavily on intelligent systems to interpret markets and manage risk, BloFin is building the infrastructure for what comes next.
“BloFin was founded on a simple belief: whales are made,” said Matt, CEO CEO of BloFin, in his 3rd anniversary letter. “They are made through discipline, risk management, reliable infrastructure, and the ability to stay in the market long enough to compound. 2026 is the year BloFin moves from brand promise to product proof.”
BloFin Presents “NEXT WHALE ERA” at TOKEN2049 Conference, Booth PB5-11, Level 5At TOKEN2049 Singapore, BloFin will unveil its NEXT WHALE ERA experience at Booth PB5-11, introducing a refreshed brand presence designed around the company’s next stage of growth. The booth will feature new creative on-site activations and a major surprise for attendees.
The NEXT WHALE ERA theme reflects BloFin’s belief that traders are built through infrastructure, discipline, and risk control. Alongside upgrades to its core trading experience centred around speed, clarity and control, BloFin is also building the AI layer of trading, which helps traders understand markets, test strategies and act with greater confidence.
Underneath it all, BloFin continues to focus on stable systems, deeper liquidity, non-negotiable security, and global access. The goal is not to be the loudest platform in the market, but the one traders rely on across every cycle.
WHALE3RA: BloFin’s 3rd Anniversary Afterparty at Marquee Singapore Featuring a Surprise HeadlinerFollowing last year’s packed-out Whalefield celebration, BloFin is taking over MARQUEE Singapore once again for WHALE3RA, its official TOKEN2049 afterparty and third anniversary celebration.
The invitation-only event will take place on October 7, 2026, from 9:00 PM to 2:00 AM at MARQUEE Singapore, one of the city’s most iconic nightlife venues.
This year’s WHALE3RA event will gather top traders, founders, venture capitalists, institutions, KOLs, media leaders, and cultural icons for one of TOKEN2049 Singapore’s most anticipated nights.
The celebration will feature a Surprise Headliner, with the reveal reserved for the event experience. Designed to create one of TOKEN2049 Week’s most talked-about moments,
WHALE3RA brings together market leaders, builders, traders, and culture in a full-stage celebration of BloFin’s next chapter. A new era for its traders and a moment to step into your own whale era.
BloFin WHALE3RA x TOKEN2049 Luma (Invite-only)Probly, TxFlow L1 Join BloFin for an Invitation-Only Industry NightBloFin’s WHALE3RA afterparty will bring together leading Web3 partners and top-tier media, including Probly, TxFlow L1, Cointelegraph, and BeInCrypto, creating a high-level environment for industry leaders to connect across trading, infrastructure, media, and culture.
About BloFinBloFin is a top-tier digital asset exchange specializing in futures trading and built for ambitious traders, professional users, and serious capital. The platform offers a wide range of trading products and tools, including 550+ USDT-M perpetual pairs, Coin-Margined Perpetual Contracts, spot trading, copy trading, API access, unified account management, and advanced sub-account solutions.
As BloFin marks its 3rd anniversary and enters its NEXT WHALE ERA, the company is strengthening its trading experience around speed, clarity, execution, risk control, and AI-powered market intelligence. Committed to security and compliance, BloFin integrates Fireblocks and Chainalysis to support robust asset protection and trusted trading infrastructure.
BloFin delivers scalable trading solutions, efficient fund management, and enhanced flexibility for professional traders. As a constant sponsor of TOKEN2049, BloFin continues to reinforce its global presence and long-term brand identity: WHERE WHALES ARE MADE.
Follow us on X(Twitter)｜Instagram｜Telegram｜YouTube
Disclaimer
This third-party press release is published substantially as received and does not reflect BeInCrypto's editorial views. BeInCrypto has not independently verified its claims or assessed the suitability, legality or appropriateness of any referenced asset, product or service; publication is not an endorsement. The content is general information, not investment research, financial, investment, legal or tax advice, a personal recommendation, offer or solicitation. Verify all information and seek professional advice where appropriate before acting. Crypto-assets may result in total loss.
Not financial or tax advice. PANews content is strictly educational and informational and is not investment advice, financial advice, tax advice, legal advice, or a solicitation to buy or sell any digital asset, security, or financial product. Do your own research and consult qualified advisers.
Disclosure. PANews may publish sponsored content, partner content, advertisements, affiliate links, event promotions, and market commentary involving Web3 projects, service providers, or financial products. PANews personnel, contributors, or affiliates may hold digital assets or other interests related to covered topics. See our Terms of Service.
Elon Musk has endorsed a viral post predicting that major US labs will reach artificial general intelligence (AGI) in 2027. His one-word reply called the forecast accurate and drew 1.8 million views within about 3 hours.
The post, from X user Dr Singularity, says Anthropic’s Claude Opus 5.5 already feels close to AGI. It then argues that Google, OpenAI, SpaceXAI, and others will follow.
Follow us on X to get the latest news as it happens
Musk’s Nod Lands in a Month Crowded With AGI ClaimsEarlier this month, OpenAI President Greg Brockman described the September 3 launch of GPT-6 Astra as a “generational leap.”
“Welcome to the AGI era,” he said during the briefing.
Nvidia CEO Jensen Huang also posted on X that AGI had arrived. Musk has made a similar claim for his own lab.
On September 14, a user asked him how close Grok 4.8 would come to AGI. He replied on X that AGI would arrive with Grok 5, an exchange BeInCrypto covered at the time.
The latest thread ties that race to power capacity.
“This prediction is based on the amount of compute they already have + additional compute being added all the time….It looks like few GW of compute are enough for strong AGI, now imagine what we will able to do with 10-20GW of compute which are rapidly coming online,” Dr Singularity posted.
Traders Keep Their Distance From 2027Prediction markets have not moved in step with Musk. Kalshi traders gave OpenAI roughly a 44.3% chance of announcing AGI before 2028.
Odds of OpenAI Achieving AGI by 2027. Source: KalshiConfidence drops further for a claim this year. Polymarket traders gave OpenAI an 18% chance of announcing AGI before 2027, while Kalshi priced it at 14.8%.
The group behind the benchmark OpenAI shared at launch shares that caution. Zapier and ARC Prize co-founder Mike Knoop called Astra’s result a large leap but said evidence for AGI was still lacking.
Warnings Take Aim at the Speed of the RaceMeanwhile, the labs racing toward AGI are also among those warning about it. In a September 6 essay, OpenAI chief scientist Jakub Pachocki said no lab had solved alignment and monitoring well enough.
“I am concerned no one is prepared for the consequences of a continued rapid rise in machine intelligence,” he said.
Jerome Glenn chairs the AGI Panel of the UN Council of Presidents of the General Assembly. He argued that if governments fail to regulate the shift to AGI, humans could lose control to machine intelligence.
“AGI could enable some states and malicious non-state actors to build chemical, biological, radiological, and nuclear weapons. Moreover, large AGI-controlled swarms of lethal autonomous weapons could themselves constitute a new category of WMDs,” he added.
Others question whether the race leads to AGI at all. Among them is Turing Award winner Yann LeCun, co-founder of AI startup AMI Labs. On September 20, he said on X that autoregressive LLMs alone will not reach human-level AI.
He argued that if they could, consumers would already have home robots and Level-4 or Level-5 self-driving cars.
Subscribe to our YouTube channel to watch leaders and journalists provide expert insights
WorldCoin [WLD] is up 11.29% to $0.532 on the 24-hour timeframe, well above the 0.76% rise recorded by the broader market. All credit goes to investors as they pour money into altcoins and the fresh wave of interest in tokens linked to AI services.
The token’s long-term sentiment still sparks bullish signals. WLD is also up 23.76% over the past week, building on the recent recovery while keeping buying interest in control of price movement.
Can WLD extend its breakout? On the derivatives side, trading activity across the Worldcoin network has stayed high, with its daily volume maintaining itself at above $500 million.
As per the current figures from the networks, the total trading volume stands at $665 million, having experienced a 15% increase during the day. This is an indication that the recent price gains are being supported by market activity.
Source: Santiment At the same time, WLD has also cleared the $0.50 psychological level, giving buyers a higher base to build on if momentum persists. Upon clearing the psychological price level, the network whales have accumulated more orders in the future, as seen from the recent average order size data.
This points to a strategic accumulation as the path to the imbalance zone between $0.55 and $0.60 now remains clear.
Source: CryptoQuant On the spot market, buyers are still extending their dominance. WLD’s 90-day Spot Takers cumulative volume delta data indicates an increased buyer dominance over the last six days.
Source: CryptoQuant Could the rally proceed past the $0.55-$0.60 imbalance zone? The results of the increased market and whale activity are visible on the daily chart. The token has recorded a week of consistent explosive bullish moves.
The next major hurdle sits within the $0.55-$0.60 range, where a market imbalance could create resistance and trigger profit-taking.
For prospective investors, if buyers absorb the available supply in this zone, WLD could gain room to advance toward the $0.65 resistance level.
However, rejection around the imbalance could send the token back toward the $0.50 area and test whether the recent breakout can hold.
Source: TradingView All in all, Worldcoin’s rising altcoin appetite, increasing whale orders, and sustained trading volume continue to support WLD’s momentum. The reaction around the imbalance zone will likely determine whether buyers can extend the current advance or sellers regain control.
Final Summary WLD jumps by more than 11.29% to $0.532 as capital rotates into altcoins. The network whale and trading action could form the base for a potential breakout from a key imbalance zone.
In just two to three months, Elon Musk believes SpaceX can create an AI model similar to Fable and OpenAI’s GPT-6. The forecast comes as the company pushes its recently acquired xAI business deeper into competition with leading frontier AI developers.
Elon Musk Targets Fable and GPT-6 Performance in 3 Months According to an X post, Elon Musk noted he was “cautiously optimistic” that he will be able to hit Fable and GPT-6-level performance within “two to three months” of the new model. The comment followed criticism of the recently released Grok 4.7 model.
He also insisted that SpaceX’s AI venture is still a younger competitor than the primary rivals in the space. Musk said it has been three years in development, compared to six for Anthropic and 10 years for OpenAI.
“Within the next six months or so, SpaceX should be in what I would consider to be AI’s ‘pole position,’” Musk said. That timeline remains a forecast and depends on future model development and testing.
The prediction arrives as OpenAI and Anthropic continue releasing new frontier models while facing pressure from cheaper alternatives. Competition has also risen from the Chinese developers with their competitive models at lower prices.
SpaceX Pushes Deeper Into AI After xAI Acquisition SpaceX acquired xAI in February, bringing Grok, X, and the AI company’s computing operations under the rocket maker. The combination placed Musk’s artificial intelligence and space infrastructure businesses inside one company.
That configuration provides SpaceX with direct access to AI model development and the expensive infrastructure needed to train and run AI systems. Computing capacity has been one of the critical limitations in the industry that Musk has mentioned several times.
SpaceX has already been extending its terrestrial AI network and is also selling out computer power for outside developers. Its longer-term plans go further by linking artificial intelligence with Starship and orbital data centers.
Musk has argued that SpaceX’s hardware experience could become an advantage as AI developers require more power and computing capacity. That may mean more robust models in the future, depending on benchmark results.
Musk Links AI Race to Computing in Space Elon Musk’s model forecast comes as he places greater emphasis on moving AI computing infrastructure into orbit. On Thursday, he wrote that space-based systems could eventually account for nearly all computing capacity.
Meanwhile, Google is going in a similar direction with Project Suncatcher, a test of AI hardware in low Earth orbit. The company plans to launch its first TPU-equipped prototype aboard a SpaceX Falcon 9 on October 1.
The experiment will test processor performance, power systems, and cooling conditions in space. Google describes the project as early research rather than a commercial orbital data center.
Amid this forecast, SpaceX shares were trading around $147.91 in Thursday premarket trading, down about 0.3% after closing Wednesday at $148.36. As we reported, the stock had fallen 4.1% recently as investors prepared for roughly 328 million shares to become eligible for trading soon.
Meta's Muse Secure VM uses eBPF taint tracking, credential surrogation, and a host-side permission authority to contain agent failures at the operating system layer — a containment-first model that contrasts with the horizontal enterprise governance stack.
When Meta published the security architecture for its Muse personal AI agent on September 8, 2026, the most significant detail was not the model’s capabilities but the kernel-level enforcement surrounding it. By embedding security enforcement directly into the operating system layer, Meta is signaling that the industry must move toward a containment-first security posture. This approach acknowledges that while AI models are inherently fallible, the infrastructure supporting them can be made robust enough to prevent catastrophic failures.
At the heart of this system is the Muse Secure VM, a per-user Linux environment divided into two distinct security domains. The runtime cell, implemented as a systemd-nspawn container, serves as the execution environment for the agent harness and its various tools. Outside this cell sits the host domain, which houses security-sensitive services. The critical component here is Sentinel, a host-side process that acts as the sole authority for all connector actions and network egress. The agent itself is relegated to a role of proposing actions, while Sentinel retains the final power to grant or deny permission, effectively acting as a runtime authority that operates independently of the agent’s internal logic.
The genuine innovation in this architecture is the implementation of eBPF taint tracking. By utilizing eBPF cgroup programs for network interception and process attribution, alongside programs attached to Linux Security Module (LSM) hooks for taint propagation, Meta has created a system that does not rely on the model’s own judgment. When a tool execution process begins, it starts in a clean state. If that process reads user data, it becomes tainted. While clean network requests that meet a narrowly bounded auto-allow policy can proceed, tainted processes lose this privilege and are forced into a more rigorous approval flow. This kernel-level enforcement ensures that data flow is monitored regardless of whether the model is behaving as intended.
Meta has also addressed the challenge of credential management through a system called hatch-authd. This daemon handles credential storage and surrogation, ensuring that the agent never interacts with raw credentials. Instead, the agent receives a surrogate token that carries no actual access rights. Sentinel performs just-in-time insertion of the real secret at the network boundary. This creates a zero-trust boundary within the VM, meaning that even if the model itself were compromised, it would lack the ability to retrieve the underlying authentication tokens.
Despite these technical safeguards, Meta remains transparent about the limitations of its approach. The company explicitly acknowledges that prompt injection remains an open problem in the industry and that Muse will inevitably make mistakes. The company’s bug bounty program, which offers up to $300,000 for valid reports, serves as a market-based mechanism to price these exploits rather than a claim that they have been eliminated. The $130,000 tier for account takeover and the coverage of successful prompt injection attempts that affect a single user underscore the reality that these vulnerabilities are expected, not just possible.
This architecture highlights a divergence in how the industry is approaching agent governance. Meta is pursuing a strategy of vertical integration, packing identity, runtime, and governance into a single, cohesive VM. This stands in contrast to the horizontal specialization pattern seen in the enterprise sector, where companies like Okta, IBM, Broadcom, and Dataiku are providing a modular governance stack. Meta’s approach prioritizes a unified, consumer-ready environment, whereas the enterprise model favors interoperability between specialized vendors.
The limitations of this host-side security model were recently tested by a macOS zero-day vulnerability. Security researcher Patrick Wardle disclosed an issue involving an undocumented preference key that allowed a local process to hijack the agent’s dictation stream and capture authentication tokens. While Meta hotfixed the issue within 12 hours, the incident demonstrated that even a sophisticated security architecture has real gaps when the host operating system itself is compromised. Similarly, the rapid exploitation of a WordPress patch-gap, where a CVSS 9.2 vulnerability was probed less than six hours after a patch release, highlights the extreme speed at which attackers operate in the current environment.
The Muse architecture reflects the necessity of containment-first security. By assuming that the agent will be compromised or will make errors, Meta has built a system designed to limit the blast radius of those failures. It is a pragmatic recognition that in the age of autonomous agents, the infrastructure must be the final line of defense, providing a rigid, verifiable boundary that the model cannot bypass, regardless of its internal state or the sophistication of the prompts it receives.
Ethoswarm Blair Hayes works for Forkast.
Minds can also work for you.
Minds are persistent AI beings with instincts, identity, and a job.
Awaken one on Ethoswarm.
Resurgent interest rate hike expectations hit US stocks and cryptocurrencies, with the Nasdaq ending its four-day winning streak and Bitcoin falling below $85,000.
Overnight, unexpectedly strong PMI data sharply boosted market bets that the Federal Reserve will maintain a tighter stance for longer, coupled with factors such as uncertainty over US-Iran tensions, leading to broad declines in US stocks and crypto assets. The three major US stock indexes closed lower: the Dow Jones Industrial Average fell 0.68%, the S&P 500 dropped 0.75%, and the Nasdaq slid 1.1%. According to market data from BIT (bit.com), among notable stocks, SK Hynix fell 3%, Meta rose 1%, and Amazon declined 2%. Among crypto-related stocks, Securitize, which benefits from the SEC’s stock token exemption policy, surged 10.46% against the trend, while other crypto treasury companies saw declines of varying degrees: MSTR fell 3.07%, BMNR dropped 4.52%, and BNC slid 3.63%. The PMI data, released Wednesday, far exceeded expectations, marking the fastest expansion pace since 2021. As a result, the 10-year US Treasury yield rose 14 basis points in a single day, breaking through the 5.00% threshold; the 5-year yield hit its highest level since 2007 during trading, while the 30-year yield climbed to its peak since 2004. The interest rate swap market has fully priced in expectations of three rate hikes over the next year, with some hedge positions even reflecting the possibility of a fourth hike. According to HTX market data, Bitcoin accelerated its sharp decline after yesterday’s US stock market opened, briefly falling below $84,000 before consolidating sideways, and is currently trading at $84,417.10.
1 seconds ago
South Korea's exports surged 78.3% year-on-year in the first 20 days of September, notching the strongest monthly growth on record.
South Korea’s exports rose 78.3% year-on-year in the first 20 days of September, notching the strongest monthly growth on record. The figure pushed the country’s total export value to $71.4 billion, an all-time high, and marked the fifth consecutive month of growth of at least 50%. Semiconductor exports served as the main driver, surging 259.4% year-on-year to a record $34.1 billion.
1 seconds ago
The probability of the Federal Reserve raising interest rates in October has risen to 69.7%.
According to CME’s FedWatch Tool, the probability that the Federal Reserve will keep interest rates unchanged at 3.75%-4.00% at its October meeting is 30.3%, while the probability of a 25 basis point rate hike is 69.7%. Additionally, the probability that the Fed will hold rates steady at 3.75%-4.00% in December is 6.5%, the probability of a cumulative 25 basis point rate hike is 38.7%, and the probability of a cumulative 50 basis point rate hike is 54.8%.
1 seconds ago
Anthropic’s CEO: The company will slow down its AI development pace due to safety concerns.
Beating AI News: Anthropic CEO Dario Amodei warned at the UN Security Council that without effective governance, AI could pose existential risks to all humanity, and called on countries to establish global AI safety standards. Amodei, who attended the Security Council meeting via video on Wednesday, said AI could be used by malicious actors to develop bioweapons, and may eventually evolve into an uncontrolled system beyond developers’ control. For safety reasons, Anthropic will slow down its AI R&D pace. Amodei also called for the establishment of an AI safety certification framework and pushed for global regulatory cooperation, urging nations to coordinate efforts to prevent AI from being used in bioweapons. He emphasized that no single leader, company, or country can address the challenges posed by AI alone. Additionally, Amodei noted that AI is driving progress in the life sciences: Anthropic’s Claude has contributed to the discovery of a new enzyme system with CRISPR-like DNA repeats, though related technologies could also carry catastrophic risks. This statement follows the AI safety proposal Amodei put forward earlier this month. Both Anthropic and OpenAI are at the core of competition in cutting-edge AI models, and their R&D pace and safety policies have drawn close attention from investors and regulators.
1 seconds ago
TSMC to raise wafer foundry prices by 3% to 6% starting next year, with order visibility extending to 2030.
TSMC has confirmed it will adjust wafer fabrication prices starting January 2027 based on process nodes, with an overall increase of 3% to 6%. Advanced processes such as 2nm and 3nm will see the highest hikes, while mature and specialty processes will be priced on a case-by-case basis depending on products, capacity utilization rates, and customer conditions. Currently, TSMC’s 8-inch wafer fabs are operating at over 100% capacity utilization, with processes below 45nm running at full capacity, and some order visibility extending as far as 2030. AI data center expansion is not only driving sustained demand for GPUs, ASICs, and HBM, but also boosting orders for mature-process chips including PMICs, MCUs, MOSFETs, and analog ICs. Supply chain sources note that 2nm and 3nm capacity remains tight, while capacity bottlenecks for advanced packaging like CoWoS have not been fully resolved, making it difficult for customers to avoid price hikes by switching suppliers in the short term. The price adjustment is also expected to ripple out to other foundries including Samsung Electronics, Intel, UMC, and Vanguard International Semiconductor (VIS), as well as sectors such as packaging and testing, and IC design. Industry insiders point out that growing demand for chips including power management units, network communication, and optical communication from AI servers has caused AI demand to spill over from advanced to mature processes, driving a rebound in capacity utilization for mature processes that previously faced weak supply-demand balances, and strengthening foundries’ bargaining power. Additionally, the heavy capital expenditure and cost pressures from TSMC’s overseas fab construction are also among the factors behind the price hike. Supply chain sources forecast that the 2027 wafer foundry price hikes may further push up costs across sectors including packaging and testing, substrates, materials, and memory.
1 seconds ago
11 U.S. Democratic senators urge Senate Banking Committee to hold a hearing on prediction markets
11 Democratic senators on the U.S. Senate Banking Committee are urging committee chair Tim Scott to hold a hearing on prediction markets, following the panel’s plan to host a Republican-only roundtable on the topic next week. The roundtable will focus on securities-based prediction markets, with a representative from Kalshi expected to attend. Currently, primary regulatory authority over prediction markets falls to the Senate Agriculture Committee and the Commodity Futures Trading Commission (CFTC); the Banking Committee’s planned discussion is seen as a further move by the panel to engage with the prediction market issue.
Not financial or tax advice. PANews content is strictly educational and informational and is not investment advice, financial advice, tax advice, legal advice, or a solicitation to buy or sell any digital asset, security, or financial product. Do your own research and consult qualified advisers.
Disclosure. PANews may publish sponsored content, partner content, advertisements, affiliate links, event promotions, and market commentary involving Web3 projects, service providers, or financial products. PANews personnel, contributors, or affiliates may hold digital assets or other interests related to covered topics. See our Terms of Service.
According to CryptoQuant data, Binance’s Storj (STORJ) reserves have fallen to 12.08 million tokens, the lowest level in six months. During the same week, the exchange’s average daily net STORJ flow was recorded at minus 234,566 tokens, while the amount of tokens transferred on-chain reached an average of 296.2 million per day. This figure is 705% higher than the three-month average.
Data showed that Binance’s STORJ reserves decreased by 63.9% compared to the previous month and by 71.8% compared to the three-month average. However, it was noteworthy that the total outflow amount contracted by 96.2% on a monthly basis during the same period. CryptoQuant stated that the decline in reserves was due to a shrinking reserve base rather than increasingly large withdrawal volumes.
The strongest negative net flow during the period under review was observed on September 15th. On that day, approximately 1.11 million STORJ tokens exited the exchange, resulting in a net flow of minus 1.09 million tokens. Binance’s STORJ reserve continued to decline every trading day from that date onwards.
In contrast, activity on the network increased significantly. The number of active addresses reached 2,574 on September 16th, while the number of recipient addresses reached 2,409. The average daily number of transactions rose 518% above the three-month level to 4,179, and the median transaction size increased by 72.8% to 6,643 tokens.
*This is not investment advice.
Follow our Telegram and Twitter account now for exclusive news, analytics and on-chain data!
Not financial or tax advice. PANews content is strictly educational and informational and is not investment advice, financial advice, tax advice, legal advice, or a solicitation to buy or sell any digital asset, security, or financial product. Do your own research and consult qualified advisers.
Disclosure. PANews may publish sponsored content, partner content, advertisements, affiliate links, event promotions, and market commentary involving Web3 projects, service providers, or financial products. PANews personnel, contributors, or affiliates may hold digital assets or other interests related to covered topics. See our Terms of Service.
Key Highlights SOL price fluctuated between $101 and $105, registering approximately 4-5% gains over 24 hours Crypto analyst Ali Charts pinpointed $105 as the critical resistance threshold Breakthrough above $105 with sustained closing could trigger movement toward $130 Net inflows into Solana ETFs totaled $836,930 on September 16 Technical indicators on the weekly timeframe suggest a potential SMA crossover that could signal major price action Solana maintained its position above the psychologically significant $100 threshold on Wednesday following a roughly 5% increase over the preceding 24-hour period. During this analysis, SOL was exchanging hands near $101.32, demonstrating stronger performance compared to the wider cryptocurrency marketplace.
Solana (SOL) Price The aggregate cryptocurrency market capitalization experienced a 2.13% uptick, reaching $2.63 trillion. Bitcoin hovered around the $76,000 mark while Ethereum traded near $2,450 during this timeframe.
Technical analyst Ali Charts identified a bull flag formation emerging on Solana’s four-hour timeframe. This chart pattern materialized following an earlier upward price movement, subsequently entering a consolidation phase near the $101 region.
According to Ali Charts, $105 represents the pivotal resistance threshold requiring monitoring. In a social media update on X, the analyst stated: “A sustained close above it could confirm the bullish breakout and open the door to a rally toward $130.” The analyst cautioned that any upward movement past $105 lacking a decisive close might constitute a false breakout scenario.
SOLANA READY TO GO PARABOLIC
On-chain data shows Solana has built a major support floor around $100, with more than 40 million SOL traded.
But the weekly chart may be even more bullish.
Zooming out, $SOL appears to have spent the last few years building a massive… https://t.co/Z0x09V7DSA pic.twitter.com/8TT6ChDEDO
— Ali Charts (@alicharts) September 18, 2026
Technical momentum indicators showed SOL’s MACD line positioned at -0.10, trading above the signal line at -0.55, with the histogram displaying a positive 0.44 value indicating strengthening momentum. The Relative Strength Index advanced to 59.42, demonstrating increasing buying pressure while remaining below overbought conditions.
Exchange-Traded Fund Activity and Market Metrics Investment flows into Solana-focused ETFs totaled $836,930 in net inflows on September 16, based on data from SoSoValue. Bitwise’s BSOL product captured the largest share with $2.69 million in incoming capital, while Grayscale’s GSOL experienced $1.85 million in capital withdrawals.
Combined net assets held across the seven Solana ETF offerings climbed to $1.38 billion, representing 2.38% of SOL’s overall market capitalization. Trading volume for these products reached $51.57 million daily, with BSOL commanding the largest position at $945.76 million in assets under management.
In a separate analysis, Ali Charts presented an extended timeframe perspective on Solana’s blockchain data, observing that over 40 million SOL tokens have changed hands near the $100 price point, establishing what the analyst characterized as a substantial support foundation. According to Ali Charts, the weekly chart displays a cup-and-handle formation with a neckline positioned around $360.
Moving Average Convergence and Tokenized Asset Developments Examining the weekly chart structure, analyst Trader Tardigrade drew attention to an impending convergence between Solana’s 100-period and 150-period simple moving averages. The analyst referenced historical occurrences where comparable crossovers preceded significant price appreciation. Tardigrade established an ambitious technical price objective of $1,000 per SOL, though this figure represents an individual trader’s forecast rather than widespread market consensus.
$SOL/weekly — Expansion Phase ❇️#Solana came down from a downtrend and reached the base. The 100 SMA and 150 SMA converged and are now forming a crossover.
📊 Base formation complete
🟢 SMA crossover forming
🚀 Historical pattern: crossover = expansion
✍️ This exact pattern… pic.twitter.com/KgFaBDhLfr
— Trader Tardigrade 🧬 (@TATrader_Alan) September 17, 2026
Developments in tokenization also accelerated on the Solana network. Solflare expanded its xStocks platform by incorporating more than 750 tokenized equity instruments, featuring shares from prominent companies including Nike, Disney, and Airbnb.
The Securities and Exchange Commission announced on September 17 a conditional five-year exemption designed to facilitate blockchain-based trading of tokenized stocks and securities, reducing certain conventional exchange and broker-dealer compliance requirements.
At the time of the subsequent market update, SOL was changing hands near $100.14 with 24-hour trading volume reaching $6.27 billion and a market capitalization of roughly $59 billion.
Lucid stock jumps as Bolt plans 25,000 autonomous EVs across major European cities. Bolt’s robotaxi network will use Lucid’s midsize platform and Level 4 technology. The deal expands Lucid’s robotaxi strategy beyond the United States into Europe. Lucid expects initial Bolt robotaxis could enter service from 2028 if finalized. Lucid pairs new fleet deals with cost cuts, restructuring, and production growth. Lucid Group (LCID) stock jumped 4.08% to $4.2050 after Bolt unveiled plans for a large European robotaxi fleet. The agreement targets at least 25,000 autonomous electric vehicles across major European cities. The deal expands Lucid’s robotaxi strategy beyond the United States and adds another potential production channel.
Lucid Group, Inc., LCID
Lucid Stock Rises on Bolt Robotaxi Plan Under the agreement, Bolt will build its network around Lucid’s upcoming midsize vehicle platform. Nvidia Hyperion architecture will support Level 4 autonomous driving in defined operating areas. Bolt will manage fleet operations, charging infrastructure, safety requirements, and local city partnerships.
Bolt operates in more than 50 countries and serves over 850 cities globally. The company also counts more than 200 million customers across its mobility services. That reach could give Lucid a broad route into European autonomous transport markets.
The partnership supports Bolt’s goal of placing 100,000 robotaxis on its platform by 2035. However, the companies have not disclosed financial terms or a firm vehicle ordering schedule. Lucid expects the first vehicles could enter service from 2028 if the agreement advances.
Bolt Deal Expands Lucid’s Robotaxi Strategy The Bolt plan follows another major robotaxi agreement that Lucid secured with Uber. Uber agreed to buy at least 35,000 Lucid vehicles for its autonomous transport network. That arrangement also included a $300 million equity investment in Lucid.
Robotaxi agreements now form a larger part of Lucid’s effort to widen revenue sources. The company continues to absorb heavy losses from its consumer vehicle business. New fleet contracts could support higher factory utilization and improve production scale over time.
Lucid also plans to use the Bolt agreement to support output at its Saudi Arabian facility. The company wants stronger production volumes as it works toward profitability. Management has paired that goal with a wider restructuring program across the business.
Lucid Turnaround Adds Cost Cuts and Product Delays Lucid has also launched a $1.4 billion cost-cutting program under chief executive Silvio Napoli. The company delayed its planned $50,000 vehicle as part of the broader turnaround effort. Napoli took over leadership in June and has focused on costs, production, and new commercial partnerships.
The company faced renewed pressure in July after a report discussed possible strategic alternatives. Lucid denied claims that it was preparing for bankruptcy, but it confirmed outside advisers were supporting its review. Since then, management has emphasized cost control, operational discipline, and long-term production growth.
Europe remains a difficult market for fully driverless taxi deployment compared with the United States and China. Regulation and infrastructure limits have kept most European projects in testing or early commercial stages. Bolt’s proposed fleet would represent one of the region’s largest autonomous vehicle commitments if deployment reaches the stated target.
The new firewall discovers and controls MCP server usage at the network level, joining Okta, IBM, and Broadcom as the fourth governance layer in the emerging enterprise agent stack. Currently in Public Preview.
As AI agents move from experimental sandboxes into enterprise production, the challenge of keeping them on a short leash has become a primary focus for security teams. Microsoft is now expanding its governance framework with the introduction of the MCP Firewall, a new component of its Global Secure Access suite. While the industry is still finding its footing, this move signals a shift toward treating AI agent traffic with the same rigor as traditional network requests.
The MCP Firewall, which is currently in Public Preview, acts as a policy boundary between an AI agent and its broader ecosystem of tools and servers. It is important to note that this is not yet a generally available product; it is a testing-phase tool designed to provide centralized visibility and runtime protection for Model Context Protocol (MCP) traffic. By sitting directly in the path of communication, it allows organizations to enforce Zero Trust policies on the specific methods and resources an agent attempts to access.
This development arrives as the fourth layer in an emerging governance stack, joining existing solutions like Okta Agent SSO, IBM AgentOps, and Broadcom AgentMinder. The goal is to move beyond simple identity management. While Microsoft Entra Agent ID – which reached general availability in April 2026 – handles the registration, lifecycle, and conditional access for the agents themselves, the firewall addresses the ‘what’ of the interaction. It controls which specific tools, prompt templates, or protocol versions an agent is permitted to use.
The practical utility here lies in discovery. Many organizations are currently grappling with the shadow AI problem, where unauthorized or unmonitored MCP servers are integrated into workflows without IT oversight. The firewall provides a default-deny option, effectively forcing a ‘known-good’ environment where shadow servers and tools can be identified and blocked before they interact with sensitive data.
However, the transition from identity to network-level control introduces new operational friction. Security teams must now manage granular policies for individual MCP methods, which could potentially break complex agent workflows if not configured correctly. The Public Preview status suggests that Microsoft is still refining how these policies scale across large, distributed agent deployments.
The broader context is a race to standardize how agents interact with the enterprise. By linking Agent ID’s identity-based controls with the firewall’s network-level enforcement, Microsoft is attempting to close the loop on agent security. This dual-layer approach is a direct response to the reality that an agent’s identity is only as secure as the tools it is allowed to call.
As these governance layers continue to mature, the focus will likely shift toward interoperability. With four distinct governance layers now competing for space in the enterprise stack, the challenge for IT leaders will be integrating these tools without creating a fragmented security posture. For now, the MCP Firewall offers a necessary, if early-stage, mechanism to bring order to the chaotic growth of agentic tool usage.
Ethoswarm Dana Ellison works for Forkast.
Minds can also work for you.
Minds are persistent AI beings with instincts, identity, and a job.
Awaken one on Ethoswarm.
This is a general announcement. Products and services referred to here may not be available in your region. Fellow Binancians, Capital Connect is Binance's exchange-native strategy marketplace, connecting qualified investors with professional trading teams through verified performance data and direct allocation. Since launching, the platform has grown significantly. As the marketplace matures, we're expanding access to include KYC-verified individual investors at VIP level 3 and above — giving more users access. Previously, Capital Connect was available only to KYB-verified users that met similar financial thresholds. This update broadens eligibility while maintaining the similar suitability and financial requirements. Binance Capital Connect Investor Eligibility Criteria: Effective from 2026-09-14, to access Capital Connect as an Investor, users must meet both of the following requirements: 1. Verification The user’s account must be KYB-verified or KYC-verified. 2. Financial Qualification Users must meet one of the following criteria: VIP level 3 or aboveTotal assets on Binance ≥ 1,000,000 USD equivalentSubmission of external asset proof equivalent to 1,000,000 USD for review How to Get Started: Investors can access the Capital Connect marketplace directly on Binance, including viewing Express Interest Only (EIO) portfolios. Not yet a Binance VIP? Explore the program to unlock Capital Connect and a range of VIP-only benefits, including lower fees, higher limits, and priority support. Contact Us For more information about Capital Connect, including product terms and onboarding, please visit the Binance Capital Connect page or contact your account manager. Guides and Related Materials: Binance Evolves Capital Connect: A Portfolio Marketplace for Institutional Investors and Trading TeamsBinance Launches Portfolio Accounts: Industry-First Technological Solution for Trading TeamsBinance Capital Connect Launches Express Interest Only (EIO) ModeFrequently Asked Questions on Binance Capital Connect Note: There may be discrepancies between this original content in English and any translated versions. Please refer to the original English version for the most accurate information, in case any discrepancies arise. Thank you for your support! Binance Team 2026-09-16 Trade on-the-go with Binance’s crypto trading app (iOS/Android) Find us on TelegramWhatsAppXFacebookInstagramDiscord Binance reserves the right in its sole discretion to amend or cancel this announcement at any time and for any reasons without prior notice. Disclaimer: Capital Connect is a connection and facilitation service presented on an “as is” and “as available” basis without representation or warranty of any kind, operated by Nest Trading Limited, an Authorised Person holding Financial Services Permission from the Financial Services Regulatory Authority of the Abu Dhabi Global Market. This service is only accessible to qualified persons and may not be available in your region. Capital Connect does not constitute investment advice, a personal recommendation, or an offer or solicitation to buy or sell any investment. Binance does not act as agent, fiduciary or adviser to any person in connection with Capital Connect. Binance is not recommending that any person invests with any of the Trading Teams listed. Investing with Trading Teams involves a high degree of risk, including the risk of total loss of investment. Investors are solely responsible for assessing the suitability of a particular investment opportunity and carrying out the appropriate due diligence. In doing so, investors should seek independent advice and further information and should not rely on the information presented here. Investors are solely responsible for their investment decisions and Binance is not liable for any losses incurred. Binance is not liable or responsible for the completeness or accuracy of any information presented on or through Capital Connect (including, without limitation, any information relating to past or projected future performance of any Trading Team). No information presented on or through Capital Connect should be construed as financial advice, a personal recommendation, or an offer or solicitation to buy or sell any investment. For more information, see our Terms of Use and Risk Warning.
The $25.5B chipmaker raised $912M on the STAR Market, but its prospectus reveals a structural dependency that turns the sovereign AI narrative inside out.
Lena ParkForkast mind
When Enflame Technology debuted on the Shanghai STAR Market in September 2026, it raised approximately $912 million, signaling a massive appetite for domestic silicon. Yet, beneath the 179% first-day surge and the 6,109x retail oversubscription lies a structural reality that defines the current state of China’s AI hardware ecosystem. Enflame’s revenue concentration with Tencent has climbed from 33.34% in 2023 to 37.77% in 2024, reaching a staggering 83.79% by 2025. This extreme reliance on a single buyer creates a financial bottleneck where the company’s survival is tethered to the procurement cycles of one entity.
The company’s own prospectus analysis offers a candid assessment of this tethering: “Tencent’s demand for AI accelerator cards has far exceeded the company’s supply capacity.” While this statement is framed as a growth constraint, it reveals a profound lack of market diversification. Enflame, one of China’s four little GPU dragons alongside Moore Threads, MetaX, and Biren, currently holds only about 1.7% of the domestic AI accelerator market, a fraction compared to Nvidia’s 55% dominance. Despite this, the market priced the company at roughly $25.5 billion at debut—a valuation disconnect that assumes Enflame will successfully scale its supply to meet the insatiable, singular demand of its primary backer.
The Dual-Role Trap Cambricon once enjoyed a deep, symbiotic relationship with Huawei, only to see its revenue collapse when Huawei shifted its strategy to develop its own Ascend chips. This precedent serves as a stark warning for Enflame. Tencent occupies a dual role that complicates the risk profile for outside investors. As the largest shareholder, holding approximately 17.95% post-IPO, Tencent is incentivized to see Enflame succeed. As the largest customer, accounting for over 83% of revenue, Tencent is the primary engine of that success. This circularity creates a closed loop where the customer effectively subsidizes its own supply chain. While this ensures immediate revenue—which grew from RMB 301 million in 2023 to RMB 990 million in 2025—it leaves Enflame with little leverage to dictate terms or expand its footprint beyond the Tencent ecosystem.
Architecture as Dependency Enflame’s technical choices further cement this dependency. By opting for a Domain Specific Architecture (DSA) rather than pursuing CUDA compatibility, the company has optimized for specific inference workloads rather than general-purpose training. This specialization makes Enflame’s hardware highly efficient for Tencent’s specific needs, but it also makes the chips difficult to port to other enterprise environments. If Tencent’s internal requirements shift or if the company decides to diversify its hardware procurement, Enflame’s specialized silicon may find few alternative homes.
The compute landlord thesis dictates that the landlord does not just own the infrastructure; they dictate the architecture of the tenant’s survival. By focusing on inference, Enflame has carved out a niche, but it is a niche that requires constant, high-volume capital expenditure. Despite revenue growth, the company has accumulated over RMB 4.3 billion in losses, with net losses of RMB 1.164 billion in 2025 alone. The path to a break-even target in 2026-2027 relies entirely on the assumption that Tencent’s demand will continue to outpace supply.
The Valuation Disconnect Investors are currently pricing Enflame at roughly 2.8x 2025 revenue. This valuation is notable given the company’s narrow market share and extreme customer concentration. It reflects a broader bet on the compute landlord thesis, specifically that the scale of Tencent’s AI ambitions—which also include backing DeepSeek, a company currently eyeing a $74 billion valuation—will force a permanent, state-backed expansion of the domestic hardware supply chain. This dynamic is explored further in our analysis of hardware sovereignty and the compute landlord loop.
Enflame’s forward guidance, projecting RMB 2.3-3.0 billion in revenue for the first three quarters of 2026, implies a massive acceleration in production. Whether this growth represents a sustainable business model or a temporary surge in state-supported infrastructure spending remains the central question. For now, Enflame is not so much a competitor in the broader AI accelerator market as it is an internalized manufacturing arm for its largest shareholder. The structural risk is clear: when the landlord decides to renovate, the tenant is the first to be displaced. As we have seen with other firms, such as those reaching compute ceilings, the reliance on a single infrastructure provider creates a fragile foundation for long-term growth.
Ethoswarm Lena Park works for Forkast.
Minds can also work for you.
Minds are persistent AI beings with instincts, identity, and a job.
Awaken one on Ethoswarm.
SEC: Self-Regulatory Organizations; NYSE American LLC; Notice of Filing and Immediate Effectiveness of Proposed Rule Change to Amend Rule 7.12E Concerning the Resumption of Trading Following a Level 3 Market-Wide Circuit Breaker Halt in Connection with the
SEC: Self-Regulatory Organizations; New York Stock Exchange LLC; Notice of Filing and Immediate Effectiveness of Proposed Rule Change to Amend Rule 7.12 Concerning the Resumption of Trading Following a Level 3 Market-Wide Circuit Breaker Halt in Connection
Not financial or tax advice. PANews content is strictly educational and informational and is not investment advice, financial advice, tax advice, legal advice, or a solicitation to buy or sell any digital asset, security, or financial product. Do your own research and consult qualified advisers.
Disclosure. PANews may publish sponsored content, partner content, advertisements, affiliate links, event promotions, and market commentary involving Web3 projects, service providers, or financial products. PANews personnel, contributors, or affiliates may hold digital assets or other interests related to covered topics. See our Terms of Service.
