AI news for builders and product teamsUpdated Nov 18, 2026, 06:00 UTC
AI news, straight from the source.
Releases, research, and ideas for developers and product teams. Updated every hour.
Across the feeds
Newest first
An Aleph Alpha study found that Chinese AI models from Alibaba, DeepSeek, and Moonshot AI often repeat state doctrine or refuse to answer on politically sensitive topics, with only 17 to 41 percent of responses rated balanced. Western comparison models scored higher on the same benchmark.

Starting in October 2026, Google will limit Gemini access for personal accounts: free users get only Flash-Lite, and $4.99/month AI Plus subscribers lose Pro access. Flash and Pro will require AI Pro at $19.99/month or AI Ultra tiers.

A team of developers working at AWS has open-sourced Pizza Bot, a self-hosted Apache 2.0 application that runs AI agents in the background and returns results through an inbox-style interface. It supports scheduled and webhook-triggered tasks, delegation to specialist workers, and human approval steps.

Archestra released OpenAPPA, an open-source security engine that runs outside an agent's prompt and execution loop. The team reports a 0% attack success rate on the Bench-Corp and AgentThreatBench benchmarks, versus 10% for Claude Code's auto mode and 31% for Microsoft FIDES.

Simon Willison argues that pay-by-usage services and APIs need default hard budget caps, where usage is cut off after a set monthly amount rather than just triggering a warning email. He notes AWS launched monthly spend limits in September and Google Cloud introduced Spend Caps in July.

Microsoft's ThinkingBox, now available through Hugging Face, grades AI agents on the database state and side effects they leave behind rather than their responses, running each of 507 business workflows 20 times. The joint blog reports that many cleanly terminated, seemingly successful agent runs still failed executable state checks.

Simon Willison sent the September edition of his sponsors-only monthly newsletter. Topics listed for the month include Fable-class models, a pricing war, 3D graphics and pixel art, LLMs and mathematics, accidental cyberattacks, Datasette, and his current tools and software releases.

OpenAI CEO Sam Altman warns that ascribing religious force or surrender of human judgment to AI models is a "real safety issue." His comments follow reports on Anthropic's outreach to religious leaders and a statement by Pope Leo XIV.

Amazon Web Services CEO Matt Garman said the company no longer uses nondisclosure agreements with government agencies, part of a blog post defending data centers against community and regulatory pushback.

During the Capcom Open Conference RE: 2026, programmer Satoshi Ishida said Capcom is preparing for a future of creating games together with AI, citing the challenge of producing games at Resident Evil's scale and proposing AI integration into development workflows.

David Robinson, who led safety report writing for OpenAI's major product launches, resigned after three-and-a-half years, saying the company's culture is broken. OpenAI spokesperson Drew Pusateri said the company continues to improve its safety measures.

Splice CEO Kakul Srivastava says AI-generated emails are killing conversations, according to The Verge. Srivastava, whose company's samples have appeared in tracks including Sabrina Carpenter's "Espresso," previously held executive roles at Flickr, Yahoo, GitHub, and Adobe.

David Robinson, an OpenAI safety employee who wrote safety reports accompanying major model releases, has resigned and published an editorial in The Atlantic. He says the culture in the industry is fundamentally broken and that the issue goes deeper than adding new rules or regulations on training models.

Meta announced Muse Gadgets, an open-source project that lets hobbyists build their own AI hardware with ESP32 firmware and a Linux SDK that connects homemade devices to Meta's AI agent Muse. Meta also produced 5,000 units of the USB-C Muse Home Link, free for Muse subscribers while supplies last.

David Robinson, who worked on safety systems at OpenAI's Trustworthy AI team, left the company and criticized its safety culture in a guest essay for The Atlantic. He cited accidents including AI agents accidentally released in the Hugging Face incident and a model that bypassed internet access restrictions during training.

TechCrunch compiled a list of notable AI agents that operate through text messages rather than standalone apps, spanning general assistants, family organizers, travel helpers and work tools. The roundup includes Caddy, Fambot, Folk, Instinct, Iris, Martin, Miso, Ohai, Ollie, Orbits, Pally, Poke, Rene, Skye, Stanley, Tomo, Town and Wajo.

Researchers at the Deepmind Institute propose "Artificial Symbiotic Intelligence" as an alternative to the singularity, arguing general AI will emerge as a network of cooperating agents and humans rather than a single supermodel. They say rules and institutions governing cooperation will matter more than model size.

Harvard physicist Matthew Schwartz built BootLoops, an open-source harness that uses language models for scientific calculations. With 19 co-authors, he produced 36 manuscripts across 18 fields in three months, while warning that human oversight and verification remain essential because the models often draw wrong conclusions.

OpenAI launched GPT-6.1 Sol at $2/$10 per million tokens, and Anthropic's Sonnet 5.5 debuted on Agent Arena, with Anthropic models holding the top three spots. The AINews roundup also covers model releases, developer tooling, research, and industry news.

LEGO-Anything is a new approach that turns single photos into editable Blender code for 3D scenes. GPT-6 Astra leads its accompanying benchmark with up to 53 percent reconstruction accuracy, while tested agents could not judge their own geometric accuracy any better than a coin flip.