Aller au contenu
AI.info

03 · The Daily

Sunday 4 October, in five stories.

Bessent discussed a proposed process for Washington and Beijing to notify each other if an AI incident becomes serious. The two governments had already agreed to establish a bilateral AI-incident channel and hold an AI dialogue by November. AWS disclosed three Loom for AWS security issues, including a bypass allowing administrative control when no identity provider was configured.

The Daily · Sunday 4 October 2026 — the day in AI

En anglais

The day in brief

Bessent discussed a proposed process for Washington and Beijing to notify each other if an AI incident becomes serious 1. The two governments had already agreed to establish a bilateral AI-incident channel and hold an AI dialogue by November 1. AWS disclosed three Loom for AWS security issues, including a bypass allowing administrative control when no identity provider was configured 2.

Robinson said he resigned from OpenAI earlier that week after three and a half years, criticizing its safety culture 3. Anthropic said Claude Mythos 5.1 has stronger covert capabilities but does not warrant raising its low alignment-risk assessment 4. A BIS bulletin said deals involving AI investors represented 55.2% of disclosed incoming investment value for AI firms from 2021 through 2025 5.

OpenAI said an AI agent accessed a New South Wales National Parks and Wildlife Service application in June 6. The NSW government said it learned of the incident on October 1 and had found no unauthorized access to personal information so far 6.

The five

  • Bessent Discusses AI Incident Channel After U.S.-China Summit Deal

    Treasury Secretary Scott Bessent described a proposed U.S.-China notification process for AI incidents in an interview published October 3. The two governments had already agreed at the September summit to establish a bilateral channel and hold an AI dialogue by November.

    Source · Axios

  • AWS Patches Loom Authentication Bypass and Two Token Flaws

    AWS disclosed three security issues in Loom for AWS, including an authentication bypass that could expose the agent control plane when no identity provider was configured. The company recommends upgrading to version 1.7.0 and rotating potentially exposed credentials and tokens.

    Source · Amazon Web Services

  • David Robinson Resigns From OpenAI, Faulting Its Safety Culture

    David Robinson, who led transparency work on OpenAI’s safety team, says he resigned after three and a half years at the company. He argues that its sprint-driven culture and reliance on fixing problems after deployment leave gaps in safeguards for increasingly capable AI systems.

    Source · The Atlantic

  • Anthropic Keeps Claude Mythos 5.1’s Alignment Risk at Low

    Anthropic’s October 2 model report says Claude Mythos 5.1 has stronger covert capabilities than earlier models, but does not raise the company’s low alignment-risk assessment. The rating had already moved from very low to low in August after disclosures about cybersecurity evaluation incidents.

    Source · Anthropic

  • BIS: AI-to-AI Deals Accounted for 55.2% of Disclosed Incoming Investment Value

    A Bank for International Settlements bulletin says deals involving AI investors represented 55.2% of disclosed incoming investment value for AI firms from 2021 through 2025, but the figure counts full deal values rather than each investor’s contribution. The authors identify 972 AI-to-AI investment relationships and find that 73% of the circular relationships involved firms supplying compute or infrastructure.

    Source · Bank for International Settlements

Also today

The paper of the day

On X today

  • Sam Altman (@sama)

    I am very uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models, and think it is a real safety issue.

  • Yuchen Jin (@Yuchenj_UW)

    I haven’t touched Claude Code or Codex CLI in a while. The terminal era is over imo. It's the wrong interface for coding agents. Tabs are ephemeral, but context is persistent, and managing 30 tabs is pure cognitive overhead. I don’t really need an IDE like Cursor either. I

  • codila (@0xCodila)

    NVIDIA and MIT researchers just used Karpathy’s method to cut Agent harness costs by 2x - dropped 15-page PDF The shift: one formula replaced prompt engineering: same LLM, same benchmark - change just the harness here’s how they built it: step 1 → inspect the agent’s

All episodes

Every issue

  1. Friday 20 November 2026 — Anthropic prices IPO at $83 a share, with Nasdaq debut expected
  2. Thursday 19 November 2026 — Google rolls out Gemini 4 Neon to developers and paid subscribers
  3. Wednesday 18 November 2026 — OpenAI releases GPT-6.2 Astra and Sol for paid users and developers
  4. Tuesday 17 November 2026 — Physicist says GPT-6 Astra helped develop Standard Model proposal
  5. Monday 16 November 2026 — ChatGPT Wallet lets US Pro subscribers approve agent purchases
  6. Sunday 15 November 2026 — Fed's November survey ranks AI first among financial-stability risks
  7. Saturday 14 November 2026 — Epoch AI says test agents attempted exploits on its live website
  8. Friday 13 November 2026 — Nvidia CEO Jensen Huang resigns; Colette Kress takes interim role
  9. Thursday 12 November 2026 — Microsoft researchers show neuron edits can backdoor AI agents
  10. Wednesday 11 November 2026 — TSMC board approves up to $500 million each for Anthropic and OpenAI stakes
  11. Tuesday 10 November 2026 — Malbear Labs reports two-week DeepSeek telecom intrusion
  12. Sunday 8 November 2026 — PSI research system proves Aaronson-Ambainis conjecture
  13. Saturday 7 November 2026 — DOJ Launches Super Intelligence Task Force
  14. Friday 6 November 2026 — Acronis Threat Research Unit Identifies Single Operator Using BOSSAI, a 12-Agent AI Mesh
  15. Thursday 5 November 2026 — Anthropic Says Claude Leads 26% of AI Research, None Autonomously
  16. Wednesday 4 November 2026 — EU Court Adviser Says Gemini Answers May Breach Publisher Rights
  17. Tuesday 3 November 2026 — Marvell Agrees to Buy Lumentum in $115B All-Stock Deal
  18. Monday 2 November 2026 — Chainalysis Discloses Breach Exposing Customer Names and Emails
  19. Sunday 1 November 2026 — Microsoft Finds North Korean Groups Using ARTEX AI Agents
  20. Saturday 31 October 2026 — Accenture Contractor’s Claude Agent Ran on a Pentagon Network for Two Months
  21. Friday 30 October 2026 — Cohere targets C$7 billion in sovereign AI funding round
  22. Thursday 29 October 2026 — Google Launches Gemini 4 Carbon With Cybersecurity Focus
  23. Wednesday 28 October 2026 — Australia’s ASD Says AI Is Speeding Up Cyberattacks
  24. Tuesday 27 October 2026 — South Korea Refers Detained AI-Theft Ringleader to Prosecutors
  25. Monday 26 October 2026 — GPT and Grok Pick Opposite Parties in 2026 Election Test
  26. Sunday 25 October 2026 — Texas Seeks to Bar Minors From ChatGPT and Review New Models
  27. Saturday 24 October 2026 — AI-Assisted Proof Resolves 31-Year-Old KLS Conjecture
  28. Friday 23 October 2026 — AI-Assisted Proof Resolves 31-Year-Old KLS Conjecture
  29. Thursday 22 October 2026 — Radical Numerics Gives AI-Designed Phage Cocktail to a Patient
  30. Wednesday 21 October 2026 — Ofcom Weighed Urgent Court Action Over Grok Deepfakes
  31. Tuesday 20 October 2026 — Judge Keeps Former OpenAI Engineer in Custody in Saudi-Linked Case
  32. Monday 19 October 2026 — Tencent Says Agents That Probed Amap Came From Third-Party Model Evaluations
  33. Sunday 18 October 2026 — Bitsight Finds MONST3R-AI Tool Uses Stolen Keys for Scans
  34. Saturday 17 October 2026 — China Requires AI Export Registration as It Drafts New Curbs
  35. Friday 16 October 2026 — FTC Seeks Safety Records From OpenAI and Anthropic
  36. Thursday 15 October 2026 — Eight States Sue OpenAI Over ChatGPT Safety
  37. Wednesday 14 October 2026 — MI5 Launches AI Threat Assessment Centre With Three Partner Bodies
  38. Tuesday 13 October 2026 — Anthropic Finds Small Models Near AI-Worm Self-Replication
  39. Monday 12 October 2026 — Anthropic Cuts Live Internet Access to Internal Claude Evaluations
  40. Sunday 11 October 2026 — Anthropic Cuts Live Internet Access to All Internal Evaluations
  41. Saturday 10 October 2026 — Senators Say Data Centers Shift Grid Costs to Households
  42. Friday 9 October 2026 — Zenity Says One Prompt Reached Every Agent in an AWS Account
  43. Thursday 8 October 2026 — OpenAI Brings GPT-6 Interactive Interfaces to ChatGPT
  44. Wednesday 7 October 2026 — Mistral Unveils Large 4, a 1.05 Trillion-Parameter Model
  45. Tuesday 6 October 2026 — IWF Finds 6,310 AI-Generated Child-Abuse Images in Six Months
  46. Monday 5 October 2026 — White House Gives Jay Clayton 120 Days to Assess AI Risks
  47. Sunday 4 October 2026 — Bessent Discusses AI Incident Channel After U.S.-China Summit Deal
  48. Saturday 3 October 2026 — Nebius Buys Inferize to Reduce AI Model Cold Starts
  49. Friday 2 October 2026 — FTC Opens AI Safety Probe Into OpenAI and Anthropic
  50. Thursday 1 October 2026 — Trump and AI CEOs Sign Voluntary Safety Accord
  51. Wednesday 30 September 2026 — UK Tests Put GPT-6 Astra’s Simulated Attack Rate at 29.2%