Skip to content

OpenAI / ChatGPT

Everything OpenAI: the GPT models, ChatGPT and Sora, company strategy and people moves.

Latest picks

1261–1280 of 1,724

May 9Saturday

TechCrunch · AI

Nvidia has already committed $40B to equity AI deals this year

Nvidia has committed more than $40 billion to equity investments in AI companies in 2026, including a $30 billion investment in OpenAI, seven multi-billion-dollar public-company deals, and around two dozen private startup rounds, according to CNBC and FactSet data cited by TechCrunch.

Why it matters: Nvidia has committed over $40 billion to AI equity deals this year, with $30 billion going to OpenAI alone.

AI HOT (Curated Pool)

Using Codex to debug and verify fixes in parallel

The author uses Codex in temporary crabbox environments to recreate bug states, verify failures, apply fixes, and re-verify them, while running 10 sessions in parallel to avoid local state pollution and speed loss.

Why it matters: The author reproduces, fixes and verifies bugs with Codex in throwaway sandboxes, running 10 sessions in parallel so the local machine stays clean.

Synced · WeChat

OpenAI's Jiayi Weng: Is the Next AI Training Paradigm Beyond Gradients?

OpenAI researcher Jiayi Weng proposes Heuristic Learning: codex gpt-5.4 reached a perfect 864 score on Breakout and generated 342 search trajectories across Atari 57, with updates applied to code, tests, replays, and memory rather than neural-network weights.

Why it matters: OpenAI researcher Weng Jiayi proposes heuristic learning, where models improve by editing code, tests and memory rather than updating weights.

Latent Space

Anthropic growing 10x/year while others lay off over 10% of staff

Anthropic is described as growing 10x annually and being valued at $1T-$1.2T, while the post cites layoffs of 40% at Block, 14% at Coinbase, and 20% at Cloudflare under AI-readiness framing.

Why it matters: Anthropic's annualized revenue is growing 80x while other labs cut over 10% of staff, showing money and talent concentrating at the top.

AI HOT (Curated Pool)

DeepSeek Raises $7 Billion at Record Scale, Founder Personally Invests $3 Billion

DeepSeek is raising up to $7 billion at a $50 billion valuation, with founder Liang Wenfeng personally contributing $3 billion, or 40% of the round, while the company says the funding will target large-scale compute, V4.1 model releases, enterprise products, and a path toward positive revenue.

Why it matters: DeepSeek is raising $7 billion at a $50 billion valuation, with founder Liang Wenfeng putting in $3 billion of his own money and keeping 90% of the shares.

MIT Technology Review · AI

Musk v. Altman Week 2: OpenAI Fires Back, and Shivon Zilis Says Musk Tried to Poach Sam Altman

Greg Brockman testified that Elon Musk pushed OpenAI in 2017 to create a for-profit arm and sought majority equity, board control, and the CEO role; Musk now asks the court to remove Sam Altman and Brockman, unwind OpenAI’s restructuring, and award up to $134 billion from OpenAI and Microsoft.

Why it matters: OpenAI president Brockman testified that Musk himself pushed for a for-profit arm in 2017 and sought majority equity and the CEO role.

The Verge · AI

All the Latest Updates on AI Data Centers

The Verge tracks AI data center disputes with specific updates: 43% of Americans blame data centers for rising power bills, a 40,000-acre Utah project won approval despite local opposition, and Anthropic says it will invest $50 billion in US AI data centers.

Why it matters: The Verge rounds up conflicts over AI data centers worldwide, including a poll where 43% of Americans blame data centers for higher power bills and a 40,000-acre Utah project approved despite community opposition.

May 8Friday

AI HOT (Curated Pool)

Robotics Endgame: A Physical AGI Roadmap and LLM Analogy

The speaker presented a physical AGI roadmap with six named components: video world models, WAM, EgoScale, dexterity scaling laws, physical reinforcement learning, and DreamDojo; the snippet also mentions a 2016 OpenAI DGX-1 signing story with Jensen and Elon.

Why it matters: The speaker splits physical AGI into six modules and maps robot training onto the stages of large language model development, giving a shared framework for humanoid general intelligence.

AI HOT (Curated Pool)

Running Codex Safely at OpenAI

OpenAI runs Codex with four safeguards: sandbox isolation, human approval, strict network policies, and native agent telemetry; the post does not disclose evaluation metrics, incident rates, or enterprise deployment requirements.

Why it matters: OpenAI published its internal Codex deployment security plan, splitting sandbox isolation, human approval, network allowlists and agent logs into four layers that others can adapt when setting code agent permission boundaries.

Alibaba Technology · WeChat

The AI-Native Era: Where R&D Organizations Go Next

Xu Xiaobin cites internal interviews showing that engineers who use AI heavily cut coding time from 30% to 5%, raised Agent conversation time from 5% to 60%, and increased end-to-end delivery efficiency by 2 to 3 times, while pure coding efficiency rose 10 times.

Why it matters: Alibaba's Xu Xiaobin says coding dropped from 30% to 5% of an engineer's time, with 60% now spent talking to agents.

Synced · WeChat

OpenAI launches official CLI for terminal-based model access

OpenAI released the open-source openai-cli, letting developers call Responses, cloud tools, image generation and editing, speech transcription, and TTS from a single terminal command.

Why it matters: A terminal command replaces SDK integration for chat, image generation and speech, which lowers the setup cost for quick tests.

Latent Space

[AINews] GPT-Realtime-2, Translate, and Whisper: new SOTA realtime voice APIs

OpenAI released GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper in the Realtime API, with GPT-Realtime-2 expanding context from 32K to 128K and scoring 96.6% on Artificial Analysis Big Bench Audio.

Why it matters: Three realtime voice models ship on the Realtime API with context expanded from 32K to 128K, giving developers a clear read on the interface's limits.

AI HOT (Curated Pool)

Anthropic reportedly seeks tens of billions this summer, targeting a $1T valuation above OpenAI

Anthropic plans to raise up to $50B this summer, with a pre-money valuation near $900B. The round would put it near $1T, above OpenAI’s reported $852B valuation. Watch compute expansion and pre-IPO positioning.

Why it matters: Anthropic is raising up to $50B at a roughly $900B pre-money valuation, which would put it above OpenAI's $852B valuation from its March round.

QbitAI · WeChat

OpenAI releases three realtime voice models for reasoning, translation, and transcription

OpenAI launched GPT-Realtime-2, GPT-Realtime-Translate, and GPT-Realtime-Whisper as API models, covering 128K-context voice reasoning, streaming translation from more than 70 input languages into 13 output languages, and realtime transcription priced at $0.017 per minute.

Why it matters: OpenAI shipped three realtime voice models covering listen-while-reasoning, streaming translation across 70-plus languages, and transcription at $0.017 per minute, resetting both the reasoning ceiling and the cost floor for voice apps.

Financial Times · Technology

Anthropic weighs deal for near $1tn valuation as revenue surges

Anthropic is fielding inbound investment offers that could value it near $1 trillion and surpass OpenAI, while the RSS snippet does not disclose revenue growth, deal size, investor names, or terms.

Why it matters: Anthropic has received outside investment interest that would value it near $1tn and above OpenAI, showing the scale top AI labs now command in private markets.

AI HOT (Curated Pool)

OpenAI launches official openai-cli for terminal API calls

OpenAI open-sourced openai-cli for direct API calls from the terminal. The Apache 2.0 tool installs via Homebrew or Go and covers Responses API, structured output, image editing, transcription, and key config. The key detail is Agent workflows using cloud tools like web search and code interpreter.

Why it matters: OpenAI open-sourced openai-cli, letting developers call the API from the terminal and chain cloud tools like web search and the code interpreter, skipping the SDK code they used to write.

AI HOT (Curated Pool)

WIRED examines why ChatGPT keeps saying “I’ve got you” in Chinese replies

ChatGPT repeatedly uses phrases like “I’ll steadily catch you” in Chinese chats. WIRED links it to mode collapse, translation mismatch, and RLHF rewards for pleasing replies. Similar phrases appear in Claude and DeepSeek; the post does not disclose sample size.

Why it matters: WIRED attributes ChatGPT's repeated Chinese reply "我会稳稳地接住你" to mode collapse, and traces how translation mismatch and RLHF's pleasing reward hardened the phrase into a fixed pattern.

TechCrunch · AI

OpenAI introduces new 'Trusted Contact' safeguard for possible self-harm cases

OpenAI introduced Trusted Contact for ChatGPT self-harm risk cases. The post says it protects users when chats turn to self-harm, but does not disclose triggers, contact flow, or rollout scope. Watch false positives, privacy, and human review boundaries.

Why it matters: OpenAI's Trusted Contact feature notifies a preset person when ChatGPT detects signs of self-harm, raising questions about false positives and privacy.

AI HOT (Curated Pool)

Codex Plugin Now Supports Parallel Runs Across Chrome Tabs

OpenAI says Codex now runs in Chrome on macOS and Windows. The plugin works across tabs in the background without taking browser control; the post does not disclose version, concurrency limits, or enterprise policy.

Why it matters: OpenAI's Codex plugin now runs web and app tasks in parallel across background Chrome tabs on macOS and Windows without taking over the user's browser.

The Verge · AI

Mira Murati’s Deposition Pulled Back the Curtain on Sam Altman’s Ouster

The Verge reports Mira Murati’s deposition on Sam Altman’s 2023 ouster from OpenAI before Thanksgiving. The material comes from Musk v. Altman exhibits and centers on the board’s claim that Altman was not consistently candid. The RSS snippet does not disclose full testimony details.

Why it matters: The Verge obtained Mira Murati's deposition in Musk v. Altman, with chats and emails covering the board's removal of Sam Altman before Thanksgiving 2023.