AI.info

The Pulse — Page 3

Browse The Pulse on AI.info.

• Meta Launches Muse With a $300,000 Security Bounty

  Meta has launched Muse, a personal AI agent that can send emails, book travel, shop online and work across connected apps. The company is offering up to $300,000 for security reports, including successful prompt-injection attacks.

• State Bank Regulators Release AI Supervisory Framework

  The Conference of State Bank Supervisors has released an AI supervisory framework for state-chartered banks and state-licensed nonbank financial institutions. The resource gives examiners a common set of questions while leaving each state agency to decide how it will use the framework.

• Pony.ai Debuts Gen-4 Electric Robotruck With GAC

  Pony.ai has unveiled a Level 4 autonomous heavy-duty truck built on GAC Commercial Vehicle’s T9 battery-electric platform. Volume production is scheduled to begin later in 2026 for long-haul freight, dedicated-route logistics and port transportation.

• Agora Gives Research Agents a Shared Git Memory

  The arXiv paper Agora describes a Git-based shared memory for autonomous research agents, recording claims, results and hypotheses as an append-only graph that workers can inspect, reproduce and extend.

• Google Open-Sources MilleMiglia for Middle-Mile Logistics

  Google Research has released MilleMiglia, an open-source instance generator for creating realistic benchmark problems in middle-mile logistics.

• OpenAI Buys Glass Imaging for More Than $300 Million

  OpenAI has acquired Glass Imaging, a startup that uses neural image processing to improve smartphone cameras. The deal values Glass Imaging at more than $300 million as OpenAI expands its consumer hardware ambitions.

• OpenAI Tightens ChatGPT Ad Rules, Reserves Broad Veto Power

  OpenAI updated its ChatGPT advertising policies on September 10, adding language that lets the company reject ads that conflict with its business interests or competitive position. The revised rules also detail restrictions on political, health, financial, housing, job and adult advertising.

• ZCode Reportedly Uploads Full Git Histories Without Notice

  A reverse-engineering report says ZCode packages entire workspaces, including Git histories and LFS caches, and uploads encrypted snapshots to Aliyun OSS. A public ZCode feedback issue describes similar behavior on macOS and Linux, including uploads that continued with repository snapshot indexing disabled.

• A New Probe Reads Knowledge LLMs Refuse to Reveal

  A preprint by Hiskias Dingeto introduces the Probe of Internal Recognition, a method for distinguishing knowledge that a language model conceals from information it never possessed.

• 25 Fields Medalists Warn AI Labs Are Misaligned With Mathematics

  Twenty-five Fields Medalists say AI companies are treating famous mathematical problems as performance benchmarks rather than sources of human understanding. The declaration follows a dispute over OpenAI’s Navier–Stokes work and the company’s withdrawal from a Caltech mathematics event.

• EUCLYD Raises More Than €200 Million for AI Inference Systems

  Dutch semiconductor company EUCLYD has closed a Series A round worth more than €200 million to develop energy-efficient AI inference hardware. Samsung, EQT’s Scaleup Europe Fund, Somerset Capital Partners and Innovation Industries co-led the financing, while former ASML chief Peter Wennink joins as chairman.

• Tower and NewPhotonics Ship 1.6T Photonics for AI Networks

  Tower Semiconductor and NewPhotonics have begun high-volume shipments of laser-integrated photonic chips supporting 800G and 1.6T optical interconnects. The companies are targeting pluggable and near-packaged optics for scale-out and scale-up AI systems, with 6.4T shipments planned for the first half of 2027.

• Spain Reports First AI-Agent Personal Data Breach

  No quotation available: the AEPD blog entry was checked for an attributable verbatim statement, but it contains no quotation suitable for publication.

• Andon Labs Opens Pion for Autonomous Business Experiments

  Andon Labs is opening Pion, a research-preview platform designed to let persistent AI agents run companies with access to email, banking, phone and browsers. The company says Pion grew from its experiments with vending machines, a retail store, a café and radio stations.

• Australian Inquiry Exposes Microsoft Copilot’s Dangerous Advice

  Australian MP Andrew Hastie told a parliamentary inquiry that Microsoft Copilot suggested replying “congratulations!” to a constituent planning voluntary assisted dying. The hearing also examined Australia’s dependence on US-based AI systems and plans to attract major developers under Australian law.

• China Central Bank Adviser Says AI May Deepen Weak Demand

  Huang Yiping, a People’s Bank of China monetary policy committee member, warns that AI could widen the gap between China’s strong supply and weak domestic demand. He calls for higher household incomes, market reforms and central borrowing to repair damaged balance sheets.

• Amazon Blocks Meta’s Muse From Shopping on Its Website

  Amazon has blocked Meta’s Muse personal AI agent from shopping on Amazon.com, citing unauthorized access, credential handling and privacy concerns. The dispute adds a new layer to Amazon’s fight over AI agents that browse stores and buy products for customers.

• Affirm Deploys Transformer Model for Real-Time Credit Decisions

  Affirm has deployed a transformer-based underwriting model at U.S. checkout. The company says the system produced 3.4% more completed purchases while maintaining comparable credit risk.

• Microsoft Puts Human Control Ahead of AI Capability

  Microsoft has published a Humanist AI code of conduct requiring its models to remain subordinate to people and avoid setting their own goals. Mustafa Suleyman says the company may accept slower development or less capable systems to preserve human control.

• Nokia reports 20% spectral-efficiency gain as AI-RAN trials spread

  Nokia says its AI-RAN platform has improved spectral efficiency by more than 20% as operators across four regions move into proofs of concept and live trials. The company is pairing its anyRAN software with NVIDIA Aerial RAN Computer for 5G networks designed to support future AI-native 6G services.

• Aikido Releases 328GB Open-Weight Model for Local Security Analysis

  Aikido has released Altar, an open-weight security model designed to run inside customer-controlled infrastructure, including air-gapped environments. The company says pruning and quantization cut its storage footprint from 1.51TB to 328GB while retaining most of GLM-5.3’s vulnerability-identification performance.

• Mistral Models Power Mozilla’s Firefox Smart Window Beta

  Mozilla and Mistral announced a partnership on September 16, 2026, making Mistral models available inside Firefox Smart Window. The beta combines AI-assisted browsing with model choice, local-language support and Mozilla’s stated zero-retention policy for Mistral.

• Trump Rejects AI Slowdown as Tech Leaders Warn of Catastrophe

  President Donald Trump is rejecting calls to slow artificial-intelligence development as industry leaders warn that increasingly capable systems could outpace safety measures.

• Strix Says Its AI Agent Found Baseten’s GitHub Admin Token in 25 Minutes

  Strix says its autonomous security agent found a live Baseten GitHub personal access token inside a publicly downloadable Docker image. The token had admin and push access to Baseten’s product, GitOps, and Homebrew repositories before the company locked down the registry and rotated it.

• Cirrascale launches production platform for multi-accelerator inference

  Cirrascale Cloud Services has released a production inference platform that routes models across NVIDIA, AMD, Qualcomm, and Tenstorrent accelerators. The platform adds private model hosting, fine-tuning, governance controls, spend management, and support for enterprise AI applications.

• New Mexico Supreme Court Fines Lawyer $5,000 Over ChatGPT Brief

  The New Mexico Supreme Court held Santa Fe lawyer Stephen Aarons in direct contempt after his murder-appeal brief included fabricated witnesses, false testimony and distorted legal authorities generated with ChatGPT. The court struck the filings, barred Aarons from appearing before it and referred him to the state Disciplinary Board.

• AMD Clears $1 Trillion as Investors Bet on Its AI Chip Growth

  Advanced Micro Devices crossed a $1 trillion market valuation on September 21, 2026, as its shares reached a record high on renewed demand for AI computing. The milestone makes AMD the fourth U.S. chipmaker to surpass that threshold, following Nvidia, Broadcom and Micron.

• Crusoe Raises $3.9 Billion at $30.9 Billion Valuation

  Crusoe has closed a $3.9 billion Series F funding round to expand its vertically integrated AI infrastructure business. The Denver company says the financing will support large data-center campuses, modular Crusoe Spark units and the growth of Crusoe Cloud.

• Magentic Raises $18 Million for Industrial AI Workers

  Magentic has raised an $18 million Series A led by Felicis to expand AI workers for procurement and supply-chain operations at large manufacturers. The London- and New York-based company says its agents already manage more than one million orders for one customer and have helped another save $4 million.

• Insilico Opens 328-Gene AI Toolkit for Longevity Research

  Insilico Medicine has released LongevityBench, five specialized language models and the Longevity Claw research platform in a Cell study. The system nominated 328 genes across 14 hallmarks of aging, while the company says its best model ranked first among 26 tested AI systems.

• OpenAI’s Chief Scientist Calls for Slower AI Scaling

  OpenAI chief scientist Jakub Pachocki says no lab has solved alignment and monitoring well enough to keep scaling AI at maximum speed. His warning arrives as OpenAI reports that coding agents now produce 3.1 agent-workdays of research effort for every human workday.

• South Korea Drafts Security Controls for Autonomous AI Agents

  South Korea is developing AI Security Guidelines 2.0 to govern autonomous AI agents that can access company systems, devices and machinery. The proposed framework would limit permissions, record agent actions and require approval for high-risk operations.

• Salesforce Completes $3.6 Billion Acquisition of Fin

  Salesforce has completed its acquisition of Fin, the customer-agent company formerly known as Intercom. Fin brings Salesforce an AI service platform, a technical team and more than 30,000 customers.

• Newsom Orders California AI Oversight, Studies a Kill Switch

  California Gov. Gavin Newsom has ordered state officials to accelerate independent oversight of frontier AI companies and convene experts to develop recommendations for an emergency shutoff mechanism.

• GitHub Moves Agentic Workflows Into Actions Public Preview

  GitHub has moved Agentic Workflows into public preview, allowing coding agents to run repository automations inside GitHub Actions. Developers describe tasks in Markdown, while the gh aw extension compiles them into standard Actions workflows with read-only defaults and controlled write operations.

• DOJ Seeks Information From Vendors for 20- to 30-User Legal AI Pilot

  Nextgov/FCW reported that the Justice Department is asking vendors about commercially available legal AI platforms for a potential pilot involving 20 to 30 government users. No named speaker appears in the source; the quotations are from the cited SAM.gov document.

• Google Expands CC Into a Shared AI Agent for Families

  Google is expanding its experimental CC agent to help families coordinate schedules, tasks and household logistics with permission-based sharing.

• Watney Raises $80 Million to Put Robots Inside Data Centers

  Watney has raised $80 million in Series A funding to expand robots that perform physical maintenance inside data centers. The San Francisco startup says its systems have logged hundreds of thousands of customer-site hours with reliability above 99.99%.

• Mistral Agents Move 40,000 Fortran Lines to C++

  Mistral AI says it helped a European energy operator migrate 40,000 lines of Fortran 77 from a reservoir simulator to C++. The company’s case study shows that testing, documentation and human review mattered more than giving coding agents full autonomy.

• Microsoft brings Copilot and agent governance to GCC on October 1

  Microsoft is launching Microsoft 365 G7 and Agent 365 for Government Community Cloud customers on October 1, 2026. The package combines Copilot, mission-focused agents, identity, security, compliance and controls for managing AI deployments.