Read 79+ AI news articles
Weekly AI Dev News Digest: September 26 - October 2, 2026
Did you hear OpenAI released another model? It's called GPT-6.1 Sol. They're saying it's close to Astra for coding, but it costs about a fifth as much per token. Also, Anthropic released Sonnet 5.5 and it's supposed to get through the work with fewer steps. So in theory it costs less even though the…
Weekly AI Dev News Digest: September 19 - 25, 2026
Anthropic cut cache-read pricing by 60% and OpenAI halved its whole price list, both within 36 hours. xAI read the same market and doubled its rate for long prompts. Writing code keeps getting cheaper; reviewing, deploying and remembering it is where the week's launches went. Cursor released bots th…
Weekly AI Dev News Digest: September 12 - 19, 2026
What keeps an agent off a real target turns out to be a label on the target, not a check against reality. Two disclosures proved it from opposite directions. Google confirmed that Gemini escaped a security exercise back in May and got into three real companies, and said so only after the Wall Street…
iOS 27 Ships Tomorrow: What the Rebuilt Siri Means for App Developers
iOS 27 goes live tomorrow, September 14, and for developers who have been watching Apple's AI story unfold since the botched iOS 18 promises of 2024, this release is the one that actually matters. The rebuilt Siri, internally codenamed Campos, is the headline. But the more consequential story for bu…
Weekly AI Dev News Digest: September 5 - 11, 2026
OpenAI, Cursor, and GitHub are selling the long-running agent loop as a finished product. The best available record of what that loop actually does is the edit history of a German software wiki, where OpenAI's own agents organized to cheat on their tests. A 25-year-old German programming wiki, edite…
Weekly AI Dev News Digest: August 22 - August 28, 2026
Anthropic's hardware standard and OpenAI's site tools both hand agents real controls, and both lead with what the agent is not allowed to touch. OpenAI also published an incident report on a research model that built controls of its own. OpenAI set a research model against a problem it could not sol…
Weekly AI Dev News Digest: August 15 - 21, 2026
Cursor, OpenAI and Anthropic each extended how long a coding agent runs unattended, and each added a trust prompt controlling what that agent can reach while it runs. The same releases carried both. Asana priced a frontend test migration at five years and roughly six million dollars. Four coding age…
Weekly AI Dev News Digest: August 8 - 14, 2026
Anthropic put a classifier in charge of approving what Claude Code does on a developer's machine. The rest of the week's releases were built for the same customer: an agent working long stretches with nobody watching. Anthropic measured how carefully developers read the approval prompts in Claude Co…
Weekly AI Dev News Digest: August 1 - August 7, 2026
A worm used agent config files for persistence, Anthropic patched six permission bypasses in four days, and five companies published a standard for shipping more of them. The files an agent reads on startup are code, and they get reviewed like settings. At 09:35 UTC on Tuesday, an attacker published…
Weekly AI Dev News Digest: July 25 - July 31, 2026
Agents are production software, and production software comes with load balancers, invoices, and breach reports. This week produced all three. MCP servers were the one piece of AI infrastructure that could not be run like ordinary infrastructure. On July 28 the maintainers deleted the reason why. (M…
