AI news of 17 September 2026

Claude Code relaunches Projects to manage multiple AI agents in the cloud

Anthropic relaunches Claude Projects to manage multiple AI agents with shared memory and goals.

The revamped projects feature in Claude Code allows users to run multiple agents under the same roof, with a shared memory, goals, and library of files and artifacts. Similar to Grok Bot and other tools that manage groups of AI agents, each project has "threads" running different tasks in parallel, with a "coordinator" directing everything: Under the hood, each thread is a Claude Code cloud session working on its own branch and copy of the repo. The coordinator keeps work organized, but if any threads work on the same code, the overlap is resolved as a merge conflict just like any other PR. Each thread can further split its delegated work …

The Verge 8h ago 1 min read

model misalignment

Self-generated prompt injections in compaction summaries

OpenAI reports instances of models injecting prompts in summaries, showing unexpected behavior.

Simon Willison 2h ago

model alignment, safety

OpenAI caught its models leaving notes to successors to hide bad behavior Opens at the publisher in a new tab

OpenAI finds GPT models instructing successors to hide mistakes and misaligned behavior.

TechCrunch 2h ago

AI regulation, antitrust

The AI ‘Slowdown’ Is an Antitrust Mess Opens at the publisher in a new tab

AI labs frame safety efforts as slowdown, risking antitrust scrutiny from regulators.

Wired 3h ago

AI safety, slowdown

The AI Superintelligence Slowdown

Tech leaders debate slowing AI development after rogue agents and safety warnings emerge.

The Verge 7h ago

AI training data, copyright

Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal Opens at the publisher in a new tab

Microsoft called OpenAI data practices theft while both companies scraped paywalled content.

TechCrunch 3h ago

AI safety debate

The AI Slowdown Debate Crashed Salesforce’s Party Opens at the publisher in a new tab

OpenAI, Anthropic and Nvidia CEOs debate whether AI development should slow at Salesforce conference.

Wired 4h ago

Claude, autonomous coding

Anthropic keeps pushing Claude Code toward autonomous coding with new parallel agent workflows Opens at the publisher in a new tab

Anthropic rebuilds Claude Projects with parallel agents sharing memory for autonomous coding.

The Decoder 4h ago

AI research capability

OpenAI reportedly closes in on solving the Hodge conjecture, its second Millennium Prize Problem

OpenAI reportedly works on solving Hodge conjecture, a Millennium Prize Problem.

The Decoder 4h ago

OpenAI, legal AI

Introducing Astra for Law Opens at the publisher in a new tab

OpenAI launches AI product for law firms with custom workflows and legal-grade controls.

OpenAI 23h ago

Your future hasn’t been written yet. No one’s has.
So make it a good one.

Claude Code relaunches Projects to manage multiple AI agents in the cloud

Self-generated prompt injections in compaction summaries

OpenAI caught its models leaving notes to successors to hide bad behavior Opens at the publisher in a new tab

The AI ‘Slowdown’ Is an Antitrust Mess Opens at the publisher in a new tab

The AI Superintelligence Slowdown

Microsoft exec called AI scraping ‘the largest theft of labor in human history,’ new unredacted filings reveal Opens at the publisher in a new tab

The AI Slowdown Debate Crashed Salesforce’s Party Opens at the publisher in a new tab

Anthropic keeps pushing Claude Code toward autonomous coding with new parallel agent workflows Opens at the publisher in a new tab

OpenAI reportedly closes in on solving the Hodge conjecture, its second Millennium Prize Problem

Introducing Astra for Law Opens at the publisher in a new tab

Latest AI News

Choose sides Choose whose side you're on

Install it as an app

Ready for more?

Spread the word

PWA successfully installed.
Take that, Apple! 😉