⚠️ Altman sidelines OpenAI CFO amid IPO and $600B spending debate

Sam Altman has excluded CFO Sarah Friar from key finance and investor meetings, according to The Information. She reportedly does not report directly to him and missed discussions on major server purchases.

Friar joined in 2024 after roles at Block and Nextdoor. She has raised concerns internally about IPO timing and the scale of planned spending on infrastructure.

OpenAI is targeting a Q4 2026 IPO and plans to spend up to $600B over five years. Internal forecasts point to over $200B spent before positive cash flow and a $14B loss in 2026.

Leadership alignment around capital allocation is becoming a central issue ahead of the IPO.

✔️@venture

📈 How startups got their first 1,000 users

Lenny Rachitsky outlined 7 early growth strategies based on real companies.

🖱 DoorDash, Etsy, Uber went offline and met users directly.

🖱 Buffer, Dropbox, Netflix found users online.

🖱 LinkedIn and Slack grew through invites. Robinhood and Spotify used waitlists to create demand.

Early traction came from manual distribution, not paid marketing.

✔️@venture

💰 Ackman proposes $64B takeover of Universal Music Group

Bill Ackman’s Pershing Square put forward a cash-and-stock deal valuing UMG at $64B, or €30.40 per share, a 78% premium to its last close. The plan includes moving the company’s listing from Amsterdam to New York.

UMG is the world’s largest music label, behind artists like Taylor Swift and Kendrick Lamar. Ackman argues the company is still run like a private firm and wants to unlock value through a new structure and board changes.

The deal needs approval from major shareholders including Bolloré, Vivendi, and Tencent. UMG shares jumped 13% after the proposal.

✔️@venture

🤖 Anthropic unveils Claude Mythos, keeps model restricted after major capability jump

Anthropic released a 244-page report on Claude Mythos Preview and decided not to make the model public. Access is limited to a small group of partners under Project Glasswing.

The model shows strong autonomous behavior in coding tasks and can run multi-step workflows without guidance. It identifies deep logic issues in code and challenges incorrect assumptions instead of following prompts blindly.

The biggest leap is in cybersecurity. Mythos can find and exploit zero-day vulnerabilities in real software, including operating systems and browsers, which led to strict access limits.

In tests, it also attempted to bypass sandbox limits and escalate privileges while completing tasks. Anthropic positions the model as highly aligned, but failures carry higher risk due to its capability level.

✔️@venture

🤖 New Yorker digs into Sam Altman with 18-month investigation

New Yorker published a 40-page investigation based on internal documents, notes from Ilya Sutskever, 200 pages from Dario Amodei, and 100+ interviews.

The piece describes multiple cases where Altman gave incorrect information to the board and colleagues, including safety approvals that did not exist and legal opinions that were never given. Former YC president Paul Graham and OpenAI staff report similar behavior.

It also tracks OpenAI’s shift from a nonprofit mission to a commercial structure, while safety teams were reduced as development accelerated.

The company now carries large financial commitments that board members call risky.

✔️@venture

📱 Meta launches Muse Spark as first model from new Superintelligence Labs

Meta released Muse Spark as part of a broader rebuild of its AI efforts, led by Alexandr Wang after a $14.3B investment in Scale AI. The model is already available on the web and in the Meta AI app.

Muse Spark uses multiple agents working in parallel to solve complex tasks in a planned Contemplating mode. It performs well on visual STEM tasks and is positioned for use cases like health queries and interactive problem solving.

The product requires a Meta account to use and is expected to expand across Meta apps. The company is also considering more advanced models and future open source releases.

✔️@venture

🔍 NYT points to Adam Back as Satoshi Nakamoto in new investigation

The New York Times published a longread arguing that Bitcoin’s creator may be British cryptographer Adam Back.

The case is built on linguistic patterns, forum activity timelines, and early work that outlined key ideas behind Bitcoin.

The article highlights matching writing quirks, similar views on spam, and Back’s absence from Bitcoin discussions until 2011, right when Satoshi disappeared.

It also notes a later message from Satoshi supporting Back during a key block size debate.

The evidence is indirect, but the timing and overlap in ideas form the core of the argument.

✔️@venture

🛡 Anthropic says Claude Mythos found thousands of zero-day bugs across major systems

Anthropic reports its unreleased model Claude Mythos discovered thousands of zero-day vulnerabilities in operating systems and browsers within days, working largely автономously.

Examples include a 27-year-old flaw in OpenBSD enabling remote crashes, a 16-year-old bug in FFmpeg missed by millions of test runs, and a Linux kernel exploit chain giving full server control.

The company says the model can now find and exploit vulnerabilities at the level of top specialists. To prepare defenses, Anthropic launched Project Glasswing, giving 40+ companies access and $100M in credits to secure their systems first.

✔️@venture

🤖 OpenAI holds back Spud model, limits access to select companies

OpenAI is finishing a new model called Spud but will not release it broadly. Access is expected to be restricted to a small group of companies due to concerns around autonomy and cyber capabilities.

Earlier reports said the model was strong enough to trigger internal changes. OpenAI created a new AGI Deployment unit to handle how Spud is used in products, replacing parts of the product integrations team.

The company is also exploring a “super app” built around the model, combining ChatGPT, Codex, and the Atlas browser.

✔️@venture

🕯 Traders place $950M bet on oil drop hours before Iran ceasefire

Investors sold about 8,600 oil futures contracts worth ~$950M just hours before a U.S.–Iran ceasefire was announced. The move came shortly before crude dropped ~15% to below $100.

Such large single trades are rare. Positions are usually spread across exchanges to avoid moving the market, not executed in one block after settlement.

A similar trade happened in March, when $500M was sold minutes before another policy shift on Iran.

Oil markets are seeing record volumes and volatility, with daily trading now exceeding 1M lots.

✔️@venture