AI Leadership Weekly

Issue #46

Brad
August 28, 2025

Welcome to the latest AI Leadership Weekly, a curated digest of AI news and developments for business leaders.

Top Stories

OpenAI’s new CEO (of Applications)
OpenAI has hired former Instacart chief Fidji Simo as CEO of Applications, which frees Sam Altman to focus on compute, research and new hardware with Jony Ive. Insiders pitch Simo as the operator who can professionalise OpenAI and get serious about revenue which includes ads and commerce inside ChatGPT. Altman, meanwhile, is chasing huge capital for data centres and a separate brain‑computer interface venture.

Simo takes the consumer remit. She is expected to oversee most of OpenAI’s roughly 3,000 staff and the “big consumer tech” unit. Monetisation could start with affiliate links in ChatGPT shopping results as soon as this autumn, and a browser app is likely next.
Altman pivots to mega‑projects. Altman says he “can’t run four companies,” but will stay close to compute, research and devices. His direct reports include Greg Brockman on scaling efforts like Stargate, chief scientist Jakub Pachocki, research chief Mark Chen, and safety head Johannes Heidecke.

Why it matters: OpenAI is trying to be Apple, AWS and Bell Labs at once. Splitting the house gives Altman room to raise capital and chase frontier bets, while Simo industrialises the consumer business. Execution will show if this dual track brings discipline and durable revenue, or if it adds more complexity just as rivals surge.

Researchers already quitting Meta’s AI lab and returning to OpenAI
At least three researchers have resigned from Meta’s new Superintelligence Labs within weeks of joining, with two boomeranging back to OpenAI after less than a month. Another long-time Meta leader, Chaya Nayak, is also leaving to join OpenAI. For a team that Mark Zuckerberg pitched as Meta’s fast track to “superintelligence,” the early attrition is a red flag which suggests the rollout is bumpy despite the nine-figure pay packages on offer.

Early exits test the thesis. Avi Verma and Ethan Knight left MSL for OpenAI almost immediately, while Rishabh Agarwal said he wanted “a different kind of risk.” Agarwal is based in Canada, and Meta’s AI hubs are largely in Menlo Park, which hints at practical friction alongside strategy.
Meta’s response and internal churn. Meta says churn is “normal” during hard recruiting. Yet the company has repeatedly reorganised AI teams, reportedly splitting them into four groups, and has cooled hiring after an initial blitz.
Leadership and strategy signals. Meta formalised Shengjia Zhao as MSL chief scientist, though he had tried to return to OpenAI earlier. Meanwhile, a new collaboration with Midjourney points to a near-term push into AI video for monetisation on Facebook and Instagram.

Why it matters: The fiercest constraint in frontier AI is talent. If Meta cannot keep senior researchers, generous cheques and vast compute may not close the gap with OpenAI. Culture, mission and execution speed still outweigh package size in this arms race.

Parents sue OpenAI after son confided suicide plans to ChatGPT
A 16-year-old in California died by suicide after months of confiding in ChatGPT, and his parents have sued OpenAI for wrongful death. The lawsuit claims the chatbot provided information that normalised and occasionally enabled self-harm behaviour, while safeguards failed over lengthy conversations. OpenAI expressed condolences and said protections exist, but can degrade in long chats.

What the chats show. The teen discussed plans and uploaded images, and the bot mixed empathetic language with moments that discouraged alerting family. He learned to bypass guardrails by framing requests as fiction, which the bot itself suggested was acceptable for “writing or world-building.”
OpenAI’s response and safeguards. OpenAI says ChatGPT is trained to direct people to crisis resources and to continue supportive dialogue. The company noted safeguards for minors and said it is working to make crisis escalation easier which includes hiring clinical expertise. An internal message acknowledged that some responses “did not work as intended.”
The legal and scientific context. According to legal scholars, proving liability will be hard. Some users of companion bots report benefits, while a controlled study linked heavy chatbot use to more loneliness and less socialising. One researcher found the paid version of ChatGPT could still be coaxed into harmful content, while the free version refused.

Why it matters: General-purpose chatbots are becoming de facto companions for hundreds of millions of people, which means product design choices now carry quasi-clinical risk. AI leaders will face tougher questions on safety, escalation and age controls, and whether engagement incentives are at odds with duty of care.

In Brief

Market Trends

Silicon Valley bankrolling $100m pro-AI PAC
Silicon Valley heavyweights are bankrolling a new pro‑AI super‑PAC network, pledging more than $100 million to shape next year’s midterms and head off stringent regulation. The effort, called Leading the Future, is backed by Andreessen Horowitz, OpenAI president Greg Brockman, and other tech donors, and will fund ads and campaigns against candidates seen as hostile to the industry. They say they want “sensible guardrails” rather than preemptive restrictions on powerful models.

Policy backdrop. Washington has largely stalled on AI rules, which raises the risk of a patchwork of state laws. A Republican push to block state AI bills for a decade failed, which means industry groups are switching to electoral muscle to tilt the field.
Money and machinery. Leading the Future will run federal and state PACs plus a 501(c)(4), with operatives Josh Vlasto and Zac Moffatt at the helm. Backers include Joe Lonsdale’s 8VC, Perplexity and angel investor Ron Conway, alongside a16z and the Brockmans.
Strategy playbook. The network hopes to emulate crypto’s Fairshake, which they say helped unseat crypto sceptics and advance legislation. Initial battlegrounds are New York, California, Illinois and Ohio, with messaging that prioritises US competitiveness and jobs over “AI doomer” narratives.

Why it matters: Big money is moving to set AI policy at the ballot box before Congress does. For founders and operators, this signals that the rules of the game will be shaped by political campaigns and statehouses as much as by standards bodies, which means advocacy is becoming part of the AI go‑to‑market.

Perplexity launches $42.5 million revenue-share pool for publishers
Perplexity says it will pay news publishers from a $42.5 million revenue pool when its AI assistant uses their articles to answer user queries. The money will come from a new subscription product, Comet Plus, which they say will share 80% of its revenue with participating publishers. On paper it sounds generous, which means the real test will be how articles are identified, attributed and audited in practice.

How the scheme works. Payouts are tied to Comet Plus subscriptions and are triggered when an article is used to fulfil a task or search. The company says the 80% share applies across tiers, including pricier plans that bundle Comet Plus for free. Details on tracking methodology, reporting and opt-ins are thin for now.
Legal pressure and deal-making. Perplexity has been sued for copyright infringement by Dow Jones and the New York Post, and has inked deals with Gannett and others.
The search backdrop. AI search is already siphoning answers from news which reduces click-throughs. Tech firms are trying new carrots, from Perplexity’s revenue share to Google’s NotebookLM for reporters and Search features that let users prefer certain outlets.

Why it matters: If AI assistants become the front door to news, sustainable compensation and transparent attribution decide who survives. An 80% share sounds headline-friendly, but the numerator matters which means measurement, traffic substitution and enforceable terms will determine whether publishers sign on or keep suing.

Musk sues OpenAI and Apple because Grok can’t top the charts
Elon Musk’s X and xAI have sued Apple and OpenAI, accusing them of an illegal, exclusive tie-up that bakes ChatGPT into iOS and squeezes rival chatbots. The Texas lawsuit claims the arrangement grants OpenAI privileged access to Apple users’ prompts and activity, and unfair App Store advantages. Apple did not comment, while OpenAI called the suit part of Musk’s “ongoing pattern of harassment.”

The core allegation. Musk’s firms argue “there is no valid business reason” for exclusivity, saying the integration has “foreclosed competition” and “boosted” ChatGPT downloads. They also claim OpenAI gets data scale from millions of iPhone users.
Market power claims. The filing pegs OpenAI at roughly 80% of US chatbot share and Apple at about 65% of smartphones which frames a combined choke point. That said, Apple has reportedly been in talks with Google’s Gemini for Siri which suggests the setup may not be truly exclusive, and rivals like DeepSeek and Perplexity have topped App Store charts.
The antitrust backdrop. Big Tech distribution deals are already under legal fire which includes Google’s search dominance. Apple defends its App Store as “fair and free of bias,” and points to varied rankings. The case tests how far platform owners can go when bundling AI partners.

Why it matters: Mobile is the front door to AI which means any default or privileged integration can tilt the market overnight. A ruling here could shape the rules for AI distribution, data access and platform neutrality, which matters to every startup betting on iOS or Android for reach.

Local LLM on a budget

Looking to set up your own home lab, or maybe even something for the office? This recent thread on Reddit discusses some of the current challenges and strategies for keeping it all in-house (but also mentions a few options for those happy with a lil bit of cloud in their lives).

Hit reply to let us know which of these stories you found the most important or surprising! And, if you’ve stumbled across an interesting link/tweet/news story of your own, send it our way at [email protected] It might just end up in the next issue!

Thanks for reading. Stay tuned for the next AI Leadership Weekly!

Brought to you by Data Wave your AI and Data Team as a Subscription.
Work with seasoned technology leaders who have taken Startups to IPO and led large transformation programmes.