A day in AI

Wednesday, 4 November 2026

40 stories · 10 from multiple sources

▲ 3 signals · 2 sources

OpenAI to acquire Pierre Computer Company, Inc.

OpenAI is acquiring Pierre Computer Company to expand capabilities for code building, understanding and review with Codex.

businesstools openai.com ·
▲ 4 signals · 3 sources

British fintech Revolut slashes AI coding costs 10x by going self-hosted

Fintech company Revolut reduced AI coding expenses tenfold by switching from Claude to self-hosted models, enabling engineers to use AI more widely for debugging and task management.

infrastructurebusiness theregister.com ·
▲ 2 signals · 2 sources

Almost no one uses AI agents to book their Uber, company says

Uber reported that only 4 people booked rides using AI agents during its busiest week last quarter, showing minimal adoption of the feature.

products techcrunch.com ·

More stories

▲ 3 signals · 2 sources

OpenAI's NEW GPT-ADA! GPT-6.2 Astra is Fable 5.5 Level?! Hy4 IS OUT, DeepSeek V5, & More! AI NEWS

Tencent announced Hy4, an open-weight model coming soon that ranks among the best open models, performing close to Claude Opus 5.5 and Kimi K3.1 Pro on computer use tasks. OpenAI is reportedly developing new models including GPT-6.2 Astra.

modelsopen source youtube.com ·
▲ 2 signals · 2 sources

Uber only fulfilled 4 rides from AI agents last quarter

Uber fulfilled only 4 rides from third-party AI agents last quarter, out of 320 million total rides, while around 500 Uber Eats orders were fulfilled by AI agents during Halloween week.

products theverge.com ·

How Figma Is Building Self-Serve Analytics with OpenAI

Figma is piloting self-serve analytics with OpenAI that lets marketers query campaign performance and conversion data without waiting for dashboards to be built.

productstools youtube.com ·
▲ 3 signals · 3 sources

Y Combinator’s Garry Tan uses AI agents so much, he spends up to $200K a year on tokens

Y Combinator's Garry Tan spends up to $200K yearly on AI agent tokens and sometimes barely sleeps because of his intensive use, he disclosed on Walter Isaacson's podcast.

toolspeople techcrunch.com ·
▲ 2 signals · 2 sources

Progress on provable inference for large language models

Researchers shared progress on cryptographic proofs for verifying that large language model inference ran correctly without revealing the model or inputs.

researchsafety anthropic.com ·

A thousand fewer paths to root

OpenAI built an autonomous pipeline that discovered and generated patches for 948 local privilege-escalation vulnerabilities in the Linux kernel.

safetytools openai.com ·

Also today

DevRev delivers autonomous self-service with GPT-6.1 Sol

openai.com

RTX 5090 Street Prices Top $9k as Nvidia GPU Costs Climb

aroundprague.cz

datasette-agent 0.5a0

simonwillison.net

datasette-llm 0.1a10

simonwillison.net

datasette-apps 0.3a0

simonwillison.net

llm 0.37.1

simonwillison.net

Everything is an AI scam now

davidralphlewis.co.uk

Qualcomm Cloud AI 100 Ultra – 150W AI Accelerator Card with 128GB at 548GB/s

qualcomm.com

Ripple-65B Preview: U.S. Open-Weight SOTA on Agentic Coding

ricursive.com

Benchfmt.rs: Bringing Go's benchmarking and benchstat workflow to Rust

bwplotka.dev

Agile Deliberation: How AI helps humans define the gray areas

research.google

Dharma Field Extractor: Any Document In, Structured Fields in JSON Out

huggingface.co

Launch HN: Kino AI (YC S23) – AI media manager and video editor

news.ycombinator.com

[AINews] Behind Latent Space: Kill My SaaS, Jev & AI in 2026 | AIEi Shanghai

latent.space

Deepseek alone accounted for 53% of all (27 trillion) tokens on Openrouter yesterday. Open-weights labs totaling over 75% now

reddit.com

Mass attempted replication of old findings of LLMs with new models

zou-group.github.io

ricursiveAI/Ripple-65B-Preview · Hugging Face

reddit.com

Use decision models with Amazon Bedrock AgentCore Evaluations

aws.amazon.com

Kimi K3.1 is likely benchmaxxed (with some proof)

reddit.com

MIT Researchers Discover AI Can Rewrite Its Own Weights

youtube.com

LFM3-Omni-2B running fully offline on my iPhone 18 Pro Max at ~80 tok/s, powered by custom WebGPU kernels in a React Native app.

reddit.com

Google will let schools disable its AI tools, but not without sacrificing Search

techcrunch.com

The Definitive Guide To Editing Video With Claude

youtube.com

Aligning Language Models with General Preferences: An Empirical Study of Direct Nash Optimization

arxiv.org

Suppressing Misaligned Behaviors in Language Models, Durably

arxiv.org

Slow is smart and smart is fast

reddit.com

LoopVQ: Sub 1-bit KV Cache Quantization for Looped Transformers

arxiv.org

Exclusive: Anthropic reports its key #AI model, Claude, has helped unearth new clues about vaults—strange, barrellike structures, found by the thousands in anim

scim.ag

Knowingly Wrong: Training LLMs Against Their Own Beliefs can Induce Misalignment

arxiv.org

Pocket-sized Linux AI box runs on 256-TOPS Axera chip

reddit.com

AI PACs spent millions trying to shape Congress — here’s how they fared

techcrunch.com

← All editions