AI news for builders and product teamsUpdated Nov 17, 2026, 18:00 UTC
AI news, straight from the source.
Releases, research, and ideas for developers and product teams. Updated every hour.
Across the feeds
Newest first
President Trump's October 28 proclamation labels foreign-made industrial robot imports a national security threat and directs negotiations with trading partners within 180 days, with tariffs, quotas and fees possible if no deals are reached.

Epsilon's 50-engineer Agentic Platform team built a multi-tenant UI testing service on Amazon Nova Act and Bedrock AgentCore Browser Tool, reporting 4–6x faster test authoring (5–10 minutes versus 30–90 for an equivalent 18-step Playwright script) and a 100 percent pass rate on replayed trajectories.

Simon Willison highlights Jacob Sansbury's account of Spawn, a multiplayer game world whose codebase is largely maintained by autonomous Fable-powered agents spending over $3 million per month in tokens, with changes shipping hourly and a stated one-hour window treated as the most severe incident level.

OpenAI has published an updated Model Spec organized around five core values intended to steer more capable AI toward sound judgment, respect for human agency and faithfulness to human intent.

MoEngage wired roughly 400 private REST APIs across seven AWS Regions into agent tools using Amazon Bedrock AgentCore Gateway, avoiding more than 20 separate MCP servers and estimating about six months of engineering saved. New APIs now onboard in hours via a spec annotation rather than days of networking and review.

Alphabet's Q3 2026 results showed revenue of almost $128 billion, up 25 percent, but earnings per share of $2.12 fell well short of the $3.02 analysts expected and the stock dropped nearly 12 percent after Google said AI Mode queries earn less than traditional search.

Dream, founded by two former Palantir executives, has emerged from stealth with a $17 million seed round led by Andreessen Horowitz to build AI world models that automate building design and engineering for the construction industry.

NVIDIA's Nemotron 3.5 Super VL, a 120B-parameter open multimodal mixture-of-experts model with 12B active parameters per token and up to 1M-token context, can now be deployed on Amazon EKS, EKS Auto Mode and SageMaker HyperPod EKS using a reference implementation with NVIDIA Dynamo and vLLM.

Elorian's Glimpse visual reasoning model scored 50.48% on DecoyCount, an object-counting benchmark it co-released with Mercor, edging past the best frontier model, GPT-6 Astra at 43.17%. The benchmark tests whether models count real objects or fall for visual shortcuts.

Bank of America emailed customers urging them not to share banking login credentials with AI agents, joining Citi, Morgan Stanley and Wells Fargo in restricting third-party agent access to accounts.

Bug0 QA Agent, released by Linearbytes, is a source-available Playwright library that turns natural-language test steps into cached browser actions for faster CI regression runs, falling back to an AI agent when the UI changes. Assertions are checked each run by vision models with multi-model consensus.

LangChain's Managed Deep Agents can now run a remote browser for the agent, with tools built on Browserbase's Stagehand SDK and a handoff flow that lets a human sign in instead of giving the model credentials. On 100 WebVoyager tasks the tuned toolset passed 90% for $1.98 total.

Wabi, an AI app founded by Replika creator Eugenia Kuyda, raised $55 million at a $1.25 billion valuation, led by IVP, after repositioning itself as a personal agent that builds mini apps from prompts.

The author of the Ben's Bites newsletter says he has been pulled into over-automating his own work with personal agents and is scaling back, deleting more than 60,000 non-opening subscribers and setting five rules for a 30-day reset.

Mistral has published the open weights for Mistral Large 4, its trillion-parameter model nicknamed Le Chonk, after the Oct. 6 preview. Terminal-Bench 4.0 scores rose from 28% to 41%, and the model topped Artificial Analysis' Coding Agent Index among open models.

Anthropic has filed its S-1 to list on Nasdaq, with trading possibly starting next month at a valuation between $1.5 trillion and $2 trillion. Analyst Joey Brookhart told Decoder the filing shows $70 billion-plus in annualized recurring revenue and roughly 85 percent gross margins on inference.

Anthropic has filed an S-1 to prepare for an IPO, with share price and other figures still to come, while Google announced Gemini 4 Carbon topping the new Artificial Analysis Intelligence Index v5 but not yet made it usable.

IBM released Granite Switch 4.1.1, a backward-compatible patch that updates the Granite Speech ASR model to cover 10 languages with native punctuation and capitalization, plus an improved Decision adapter, without retraining or touching the base model.

Chinese AI labs, including DeepSeek and Moonshot AI, have shifted their training-data extraction toward OpenAI's GPT-6 Astra after Anthropic hardened Claude against such attacks, according to The Information.

S-GAME's Phantom Blade Zero launched on GeForce NOW on Oct. 28, leading a week of nine new library titles, mod.io support for Cyberpunk 2077 and The Witcher 3: Wild Hunt — Remastered, and new four-month memberships in India.