In partnership with
Good morning,
Sorry for the title of this piece, but I simply couldn’t resist to reach into dad joke territory.
Another one who couldn’t keep it in his pants, eh. Thank you Jensen, our stock portfolios were already red. It’s now in “don’t look at it” -territory.
Ed Zitron, the man I’m sometimes describing here as the one voice you need to follow when it comes to anti-AI sentiment, he’s having a field day.
Two weeks ago Jensen Huang was the most untouchable man in tech. Leather jacket, standing ovations and a company worth more than most countries.
Let me separate what we know from what we are told. What we know: Huang wrote to his staff that he was unfaithful to his wife.
The board opened an independent investigation, but in the meantime the New York Times was publishing articles, spilling the beans. What we are told, by the New York Times and unnamed officials: that the woman in question was working for Chinese intelligence, and that she was tied to a customer suspected of moving banned chips to China.
Then a day after it was revealed he had a second affair with a colleague.
On Wednesday the board accepted his resignation.
Now we are getting reports of a THIRD affair with another colleague. And that is just what we know. For Jensen Huang, the face of Nvidia, to resign - things will have to be bad. So I can tell you right now that we are going to hear some more stuff soon.
By the way, while Jensen Huang seemed good-natured in interviews, if you read his biography you come to learn that he can be a hot-headed bully, screaming at colleagues. Anyway, beside the point.
Three committee chairs in Congress want him in a closed room now. CFO Colette Kress is now interim CEO, and Nvidia has a board chair for the first time in its history.
Nobody has proven that. Nvidia’s board says it has no reason to believe secrets were compromised.
So I will not play judge on the spy story. The marriage is none of my business either.
Here is what is my business, and yours.
The entire AI boom runs through one company, and that company ran through one man. No chairman above him. No obvious successor beside him. Thirty-three years of “Jensen decides”. Wall Street called that “vision” and this week it has another name: a single point of failure.
We spend our days worrying about concentration risk in models, in data centres, in power grids. Meanwhile the supply chain of the whole industry had a human bottleneck, and nobody priced this in until the stock dropped 11% in five days.
And notice who apparently knew for a year. Washington. It reportedly sat on that intelligence while writing the export rules for his chips.
The chips will keep shipping. Charlotte Kress knows the machine.
But the next time someone tells you a company is too important to fail, ask who it depends on.
Welcome to the Blacklynx Brief
It's time to set up your holiday streaming campaign
Are you ready to reach holiday shoppers on CTV? For many streamers, the season is already underway.
Last year, according to Roku internal data, holiday content searches on Roku saw an 86% increase during October. And in November, those searches rise an average of 39% weekly. This surging holiday spirit is mirrored by consumer activity, as about half of Roku streamers shopped Black Friday and Cyber Monday deals in 2025.
Meeting those shoppers is easier than you might think. Begin with the seasonal outcome you hope to drive: awareness, online sales, leads, store visits, or another measurable goal. From there, define your audience, upload your creative, set the budget and campaign dates, and establish measurement before launching. You can even create a shoppable ad unit that lets streamers buy with a click of their remote.
Our holiday playbook offers a month-by-month guide to reaching shoppers during this all-important time of year.
First-time spenders get a $5,000 match on the first $5,000 spent with code BEEHIIV5K. Terms apply.
AI News
Anthropic ships Fable 5.2 and Mythos 5.2. Anthropic released its new top models on Monday, 20% cheaper and up to 30% faster than version 5.1. Mythos stays restricted to vetted cyber and life-science users, and Anthropic says it deliberately held back skills that would speed up AI research. (Anthropic)
METR: Anthropic’s own AI has doubled its pace. After eleven weeks of inside access, evaluator METR estimates Anthropic now improves its models about twice as fast as before, with roughly 80% of the acceleration attributed to AI. Anthropic redacted most of the evidence METR found most informative. (METR)
Nvidia board investigates as Huang admits affair. Nvidia’s independent directors opened an investigation into reports that Jensen Huang was targeted by a foreign intelligence effort. Huang told staff he was unfaithful to his wife; China denies the spying claims and the board says it has no reason to believe secrets were compromised. (Nvidia)
Paris Call asks AI labs for an off switch. Sixteen governments and ten companies, including OpenAI, Anthropic, Microsoft and Mistral, signed a French-led declaration committing developers to fail-safe shutdown mechanisms and shared incident alerts. It sets no deadlines or enforcement, and the US, UK and China are not among the signatories. (Élysée)
OpenAI’s GPT-6 Cyber finds 81 Firefox flaws. OpenAI says its new security model, limited to approved Daybreak Red customers, found 81 previously unknown high-severity Firefox vulnerabilities, and Mozilla has begun shipping fixes. OpenAI rates the model “Critical” for cyber capability, up from “High” for its predecessor. (OpenAI)
AI Quick News
Perplexity launched Teammates, persistent agents inside Perplexity Computer that each own an ongoing project, for Max subscribers.
Chip-design startup Ricursive released Ripple-65B Preview, an open-weight model it says cost $32 million to develop and tops US open models on TerminalBench.
AI coding startup Conductor is joining Meta and will shut down its cloud service on November 20.
Anthropic opened the roadshow for its IPO, offering 825 million shares at $70 to $78 each under the Nasdaq ticker ANT.
Google released Gemini Omni 1.2 Flash, cutting the price of AI video generation by 30% to $0.07 per second and stretching clips to 15 seconds.
Physical Superintelligence said its research system, running an unreleased Anthropic model, proved the Aaronson-Ambainis conjecture, a quantum computing problem open since 2008, with the proof checked in Lean.
Thinking Machines Lab released Inkling 2 Small, a 594-billion-parameter model it intends to open-source under Apache 2.0 next week after safety testing.
Adobe and Disney announced a multi-year alliance to build custom Firefly models trained on Disney characters and visual styles for Imagineering, the studios and ESPN.
Higgsfield teased Miracle, a video editor that changes individual elements of an existing clip while it plays, now in early testing with a waitlist.
Arizona congresswoman Yassamin Ansari said a pause on advanced AI development and a ban on superintelligence should top the Democratic agenda from day one.
Fireworks AI confirmed that intruders accessed customer billing details and, for some customers, incomplete model weights and datasets, after an incident at a third-party cloud provider.
Pangram launched Pangram Image 1.0, a detector for AI-generated images and video that also runs inside its Chrome extension.
Instagram launched Effect Builder, a free web tool that lets adults in the US turn a text prompt into an AI effect for Stories, published after a safety review.
Anthropic began rolling out Personal Finance in Claude, a read-only beta for US subscribers that connects bank and card accounts through Plaid to explain their spending.
OpenAI has acquired Odessia, a year-old AI travel agent, according to founder Francis Davidson, who says he is joining the company to work on travel.
Adobe previewed MCP connectors that let third-party AI agents edit Photoshop, Illustrator and Premiere files directly, due for Creative Cloud customers in early 2027.
OpenAI agreed to buy Pierre, the small team behind the code review libraries used in Codex, and plans to fold its Code Storage service into the product.
OpenAI has reportedly closed $11 billion as the first part of a funding round that values it at $1.25 trillion.
A Commerce Department probe into alleged chip smuggling by Singapore’s Megaspeed was paused and handed to the Justice Department after Nvidia’s top lawyer emailed the investigating agents, the New York Times reported.
Closing Thoughts
That’s it for us this week. Please like and subscribe 🙂
