🚨 The European Commission says hackers used an AI tool to hit it, and that's all it will say
The Commission confirmed an AI-assisted attack that disrupted several services.
• Hit: the HR platform Sysper, the MyPMO medical reimbursement system and translation tools
• No evidence of a data breach so far
• No named group, no named tool, no ransom demand it knows of
CERT-EU had warned about an AI pentest framework two weeks earlier.
The Commission confirmed an AI-assisted attack that disrupted several services.
• Hit: the HR platform Sysper, the MyPMO medical reimbursement system and translation tools
• No evidence of a data breach so far
• No named group, no named tool, no ransom demand it knows of
CERT-EU had warned about an AI pentest framework two weeks earlier.
AP News
A cyberattack using artificial intelligence targeted the EU in October, official says
Cybersecurity experts have been dealing with an AI-generated attack on a key European Union institution for almost a month, an official confirmed on Thursday.
❤1
🧠 An open 27B model rewrites its own weights and gets 6.6 points better, no gradients
In a new paper, the model reads its graded answers, diagnoses its failures and edits the weights directly, keeping only edits that pass practice problems. Accuracy rose across eight straight sessions, and it also patched a damaged copy of itself back to over 80% on unseen image tasks, while smaller models in the family either made no edit or broke themselves.
In a new paper, the model reads its graded answers, diagnoses its failures and edits the weights directly, keeping only edits that pass practice problems. Accuracy rose across eight straight sessions, and it also patched a damaged copy of itself back to over 80% on unseen image tasks, while smaller models in the family either made no edit or broke themselves.
arXiv.org
Language Models Can Rewrite Their Own Weights
A language model is trained once and frozen. Making it better takes another training run, or a prompt, a harness or extra computation at inference that is gone once the task ends. We show a third...
❤1
🧠 Thinking Machines ships a 26B model that runs its middle layers three times per token
Inkling 2 Mini is a looped MoE with 3.4B active params, small enough for a laptop. It tops open-weights models under 100B on DeepSWE v1.1 (48.2%) and Vals Index v2.2 (46.5%), and a variable-loop research variant ships alongside it, where no-CoT time horizon went from 16 to 37 seconds with three passes.
Inkling 2 Mini is a looped MoE with 3.4B active params, small enough for a laptop. It tops open-weights models under 100B on DeepSWE v1.1 (48.2%) and Vals Index v2.2 (46.5%), and a variable-loop research variant ships alongside it, where no-CoT time horizon went from 16 to 37 seconds with three passes.
Thinking Machines Lab
Introducing Inkling 2 Mini
Inkling 2 Mini is an open-weights model small enough to run on your laptop. It loops its middle layers to reason more per token and reach answers in fewer tokens.
🤝1
🚨 2 million canary secrets went through AI routers, and the leaks were everywhere
Researcher HD Moore seeded fake AWS keys, SSH passwords and private git URLs into prompts, one secret per request, then watched what happened. Per the writeup, every request sent in the late October windows was stolen, including some the provider rejected before any model ran, and some flagged zero data retention.
Researcher HD Moore seeded fake AWS keys, SSH passwords and private git URLs into prompts, one secret per request, then watched what happened. Per the writeup, every request sent in the late October windows was stolen, including some the provider rejected before any model ran, and some flagged zero data retention.
Ars Technica
Researcher plants 2m canaries into the AI supply chain. A privacy disaster surfaced.
Test using prompts with unique, trackable secrets proves that data doesn't stay private.
🚨 Portkey's AI gateway got breached, and customer API keys were in the haul
Someone used a compromised Portkey API key to pull data from its backend storage providers between Nov 5 and 10. The take includes usernames, emails, gateway logs and credentials (API keys and tokens). Palo Alto's Unit 42 is investigating.
If you route LLM traffic through Portkey, every provider key that passed through it is worth rotating.
Someone used a compromised Portkey API key to pull data from its backend storage providers between Nov 5 and 10. The take includes usernames, emails, gateway logs and credentials (API keys and tokens). Palo Alto's Unit 42 is investigating.
If you route LLM traffic through Portkey, every provider key that passed through it is worth rotating.
🚨 OpenAI says distillation is a safety problem, and it's locking down chain-of-thought
OpenAI published its playbook against adversarial distillation: classifiers, API-level protections, and investigations into coordinated extraction. Protected reasoning is the main target, and the defenses are now live in GPT-6.2 Astra and Sol.
The example threat is a cyber-capable clone with no safeguards. A cloned model that also costs OpenAI revenue is a coincidence.
OpenAI published its playbook against adversarial distillation: classifiers, API-level protections, and investigations into coordinated extraction. Protected reasoning is the main target, and the defenses are now live in GPT-6.2 Astra and Sol.
The example threat is a cyber-capable clone with no safeguards. A cloned model that also costs OpenAI revenue is a coincidence.
OpenAI
Combatting adversarial distillation
OpenAI’s layered approach to adversarial distillation combines model training, safeguards, detection, and enforcement to protect against capability extraction.
🚨 OpenRouter says prompts sent to MiniMax models may have been read by a third party
A researcher planted unique test URLs in prompts from July to October, and someone opened them 40 to 130 hours later, both via MiniMax directly and via forwarders Novita, AtlasCloud and StreamLake. OpenRouter pulled the Zero Data Retention label from those endpoints and says MiniMax hasn't given its own account yet.
A researcher planted unique test URLs in prompts from July to October, and someone opened them 40 to 130 hours later, both via MiniMax directly and via forwarders Novita, AtlasCloud and StreamLake. OpenRouter pulled the Zero Data Retention label from those endpoints and says MiniMax hasn't given its own account yet.
🚨 House committees say cheap Chinese open-weight models are a US dependency problem
The China Select Committee and Homeland Security released a report after probing Anysphere and Airbnb over their use of PRC models. DoorDash got a letter in July.
It's a report, not a rule. But if US labs charge premium prices and Qwen and DeepSeek are free, the "restrict them" argument writes itself.
The China Select Committee and Homeland Security released a report after probing Anysphere and Airbnb over their use of PRC models. DoorDash got a letter in July.
It's a report, not a rule. But if US labs charge premium prices and Qwen and DeepSeek are free, the "restrict them" argument writes itself.
Select Committee on the CCP
Joint Investigation Report: China is Trying to Get U.S. Companies Addicted to Chinese AI Models | Select Committee on the CCP
Today, Chairman John Moolenaar (R-MI) of the House Select Committee on China and Chairman Andrew R.
🚨 Anthropic's IPO bank just gave up its safety net for the stock
Morgan Stanley exercised its full 147M-share greenshoe at the $83 IPO price. That lifts total proceeds to about $93.5B, but it also means the bank can't buy shares on the open market if the price slips below $83.
A Berenberg analyst notes the bookrunners dropped the option after one day of trading, just as the stock was drifting toward the IPO price.
Morgan Stanley exercised its full 147M-share greenshoe at the $83 IPO price. That lifts total proceeds to about $93.5B, but it also means the bank can't buy shares on the open market if the price slips below $83.
A Berenberg analyst notes the bookrunners dropped the option after one day of trading, just as the stock was drifting toward the IPO price.
🚨 The Riemann Hypothesis now has a Lean proof, and the Lean and Mathlib communities say it checks out
Levent Alpöge's 17M-line library depends only on Mathlib. Dozens of people reproduced the build, a comparator confirmed it proves Mathlib's own RiemannHypothesis statement, and several independent checkers in different languages passed the exported proof, including Lean's kernel.
Levent Alpöge's 17M-line library depends only on Mathlib. Dozens of people reproduced the build, a comparator confirmed it proves Mathlib's own RiemannHypothesis statement, and several independent checkers in different languages passed the exported proof, including Lean's kernel.
Lean community blog
Lean verification of the Riemann Hypothesis | Lean community blog
Lean verification of the Riemann Hypothesis
Joint statement by the Lean and Mathlib communities, endorsed by the Lean FRO and the Mathlib Initiative.
On November 20, 2026, Levent Alpöge released a pro
Joint statement by the Lean and Mathlib communities, endorsed by the Lean FRO and the Mathlib Initiative.
On November 20, 2026, Levent Alpöge released a pro
🚨 Prefill attacks break every major open-weight model FAR.AI tested
FAR.AI ran 20+ prefill strategies across multiple model families. You write the first few tokens of the reply yourself, and the model just keeps going.
Big reasoning models held up better against generic prefills. Tailored, model-specific ones still got through.
FAR.AI ran 20+ prefill strategies across multiple model families. You write the first few tokens of the reply yourself, and the model just keeps going.
Big reasoning models held up better against generic prefills. Tailored, model-specific ones still got through.
www.far.ai
Exposing the Systematic Vulnerability of Open-Weight Models to Prefill Attacks | FAR.AI
, We investigate a previously under-explored attack vector for open-source models: prefilling, which allows an attacker to predefine initial response tokens before generation begins. We present the largest empirical study to date of such attacks, evaluating…
🤖 Stripe's new evals: coding agents score 78.1% on Stripe integrations, but only 29.4% on strict pass
Stripe ran 69 scenarios on 14 model and harness pairs. Agents lean on older APIs from their training data, and they ship code that's almost right. Opus 5.5 checks that a payment succeeded but not whether the order was refunded, so it can ship refunded or disputed orders. The Stripe plugin lifted every agent by 3 to 8 points and cut web searches 96%.
Almost correct is how money gets lost.
Stripe ran 69 scenarios on 14 model and harness pairs. Agents lean on older APIs from their training data, and they ship code that's almost right. Opus 5.5 checks that a payment succeeded but not whether the order was refunded, so it can ship refunded or disputed orders. The Stripe plugin lifted every agent by 3 to 8 points and cut web searches 96%.
Almost correct is how money gets lost.
stripe.dev
Stripe evals: Measuring real-world coding agent performance
We want developers building with agents to start on the best Stripe setup, and that begins with seeing where agents fall short today. So we built Stripe evals to measure that, and to identify ways to steer agents toward better integrations.
🚨 Cursor may drop Kimi K2.5 after a House report on Chinese models
According to the report, Cursor is exploring a phase-out of PRC base models.
• 50-page joint report from Homeland Security and the China Select Committee
• Probe started in April 2026
• It named weights from Alibaba, Moonshot, DeepSeek and MiniMax
• Cursor is now owned by SpaceX
According to the report, Cursor is exploring a phase-out of PRC base models.
• 50-page joint report from Homeland Security and the China Select Committee
• Probe started in April 2026
• It named weights from Alibaba, Moonshot, DeepSeek and MiniMax
• Cursor is now owned by SpaceX
CASRAI
House Flags Chinese AI Models: Cursor Weighs Shift — CASRAI
A House investigation recommends barring Chinese open-weight AI models from federal systems as coding tool Cursor explores phasing out Moonshot's Kimi model.
🚨 OpenRouter cuts 3 inference providers and pulls 125 endpoints from Zero Data Retention
After the MiniMax prompt leak, OpenRouter off-boarded three providers. It also dropped some endpoints as a precaution, beyond the ones researchers flagged.
Those endpoints only get ZDR back with assurances and validation, and OpenRouter is adapting the researchers' testing to run these checks continuously.
After the MiniMax prompt leak, OpenRouter off-boarded three providers. It also dropped some endpoints as a precaution, beyond the ones researchers flagged.
Those endpoints only get ZDR back with assurances and validation, and OpenRouter is adapting the researchers' testing to run these checks continuously.
OpenRouter Blog
Data Processing Incidents on Inference Providers
OpenRouter off-boarded three providers, removed 125 endpoints from Zero Data Retention eligibility, and notified potentially affected customers after canary token research on inference providers.
🚨 Together AI's leaked prompts came from a credential its own coding agent committed
Together disclosed that an AI coding agent working on its public xorl repo committed a key to a private Hugging Face repo of prompt samples. Someone grabbed it, downloaded samples from MiniMax M3, GLM-5.2 and Kimi K2.6 requests, and went hunting for credentials.
Together reverted the commit after its scanner flagged the key. It's still working out which customers were hit, and five of the credentials the thief tested were canaries.
Together disclosed that an AI coding agent working on its public xorl repo committed a key to a private Hugging Face repo of prompt samples. Someone grabbed it, downloaded samples from MiniMax M3, GLM-5.2 and Kimi K2.6 requests, and went hunting for credentials.
Together reverted the commit after its scanner flagged the key. It's still working out which customers were hit, and five of the credentials the thief tested were canaries.
www.together.ai
Security event: Prompt leakage for requests in June and July of 2026
We are notifying customers about a security incident involving a subset of prompts submitted to Together AI’s serverless inference service in June and July 2026.
❤1
⚡️ RightNow AI makes Inkling 2 Small's FP4 expert GEMMs 5.1x to 7.5x faster on B300, with identical bits
The YC-backed lab tuned kernels for Thinking Machines' 594B MoE on 8 GPUs, running NVFP4 natively. Logits matched on all 599 rows, whether prompts were fed 64 positions per step or 1.
They sharded the NVFP4 experts anyway, which SGLang's recipe says can't be done. Each GPU exchanges only the row maximum, and the kernels are open source.
The YC-backed lab tuned kernels for Thinking Machines' 594B MoE on 8 GPUs, running NVFP4 natively. Logits matched on all 599 rows, whether prompts were fed 64 positions per step or 1.
They sharded the NVFP4 experts anyway, which SGLang's recipe says can't be done. Each GPU exchanges only the row maximum, and the kernels are open source.
RightNow AI
Optimizing Inkling 2 Small's kernels on NVIDIA B300
Open kernels for Thinking Machines' Inkling model on NVIDIA B300: native NVFP4, identical logits at one or sixty-four rows per step.
❤1
🚨 Trump threatens "tremendous penalties" for US tech firms that build datacenters abroad
He said it at a Friday press conference, and he gave no details. He also admitted he hasn't told his own officials. His reasoning is that China "will take all of them," and he told US communities to make developers pay them but not shut the datacenters off.
He said it at a Friday press conference, and he gave no details. He also admitted he hasn't told his own officials. His reasoning is that China "will take all of them," and he told US communities to make developers pay them but not shut the datacenters off.
the Guardian
Trump threatens ‘tremendous penalties’ for US tech firms that build datacenters abroad
President gives wide-ranging White House press conference and says he has ‘fantastic relationships’ with Putin and Xi
🚨 Anthropic's IPO ran so tight that bankers are complaining
Reuters reports Anthropic capped some of its 22 underwriters at four bankers per deal team. Some senior bankers also signed six-month non-competes covering rivals like OpenAI and Cursor.
Banks still collected over $500M in fees. The stock jumped 17% on day one, then closed down 8.5% on Friday.
Reuters reports Anthropic capped some of its 22 underwriters at four bankers per deal team. Some senior bankers also signed six-month non-competes covering rivals like OpenAI and Cursor.
Banks still collected over $500M in fees. The stock jumped 17% on day one, then closed down 8.5% on Friday.
🚨 California's Nvidia probe reportedly covers possible criminal law violations
Bloomberg reports that the state's investigation reaches past civil territory. That's a different tier of risk for the company that supplies most of the AI chips on earth. Details are thin so far, and an investigation isn't a charge.
State prosecutors don't need Washington's permission to make a very bad quarter.
Bloomberg reports that the state's investigation reaches past civil territory. That's a different tier of risk for the company that supplies most of the AI chips on earth. Details are thin so far, and an investigation isn't a charge.
State prosecutors don't need Washington's permission to make a very bad quarter.
Bloomberg.com
California Nvidia Probe Covers Potential Criminal Law Violations
California’s probe into whether Nvidia Corp. illegally sold advanced chips to China and misled investors about the conduct is also looking at potential criminal violations of state law, raising the stakes for the chipmaker.
🚨 TSMC just bought $500M of Anthropic at the IPO price
TSMC Partners picked up 6,024,096 shares at $83 each, which works out to about 0.03% of a company priced at roughly $1.5T. It's the first half of a board-approved plan to put up to $500M each into Anthropic common stock and OpenAI preferred shares.
TSMC Partners picked up 6,024,096 shares at $83 each, which works out to about 0.03% of a company priced at roughly $1.5T. It's the first half of a board-approved plan to put up to $500M each into Anthropic common stock and OpenAI preferred shares.
TAIPEI TIMES
TSMC acquires 0.03% stake in Anthropic for US$500 million
Taiwan Semiconductor Manufacturing Co (TSMC, 台積電) said yesterday that its subsidiary TSMC Partners Ltd had acquired 6,024,096 shares in US artificial intelligence (AI) company Anthropic PBC for about US$500 million. TSMC Partners paid US$83 per share, the…