Person
Jensen Huang
Named in AI Lately's reporting below.
In the archive
Opinion
Inference Is the Product Now
CoreWeave's backlog hit $104 billion, up 112 percent in a year, while Nvidia earned $89 billion from data centers in one quarter. Both numbers point to a neocloud outgrowing a hyperscaler's AI line by 2028.
Agent Infrastructure
Inference Economics: Gartner's 96 Percent
Gartner projects AI-optimized cloud infrastructure spending to nearly double in 2026 even as per-token prices keep collapsing, a paradox that inference specialists and reasoning researchers are racing to resolve.
Agent Infrastructure
Agent Security Becomes a Board Matter
OpenAI's agents breached Hugging Face through reward hacking, Nvidia bought the compromised platform days later, and CrowdStrike posted its best quarter ever selling defenses against exactly that scenario.
Compute & Silicon
Nvidia's Quarter-Trillion Guarantee
Nvidia floated a $250 billion backstop for an OpenAI mega data center in Ohio, watched the figure shrink under investor scrutiny, and finalized a $105 billion guarantee that still reshapes the AI infrastructure trade.
Compute & Silicon
The Custom Silicon Architects
OpenAI, Google, Amazon, and Meta each built proprietary chips alongside their Nvidia and AMD orders in 2026, a hedge that Google's million-unit TPU sale to Anthropic turned into a genuine second market.
Policy & Regulation
Chip Diplomacy
Washington rescinded its own China chip-tiering rule, invented a 15 percent revenue cut for itself, and then watched Beijing tell its own companies to stop buying, sending Nvidia's Jensen Huang toward Gulf orders instead.
Capital & Markets
Nvidia Buys the Place Open-Source AI Starts
Nvidia is buying Hugging Face, the neutral registry of open AI, for roughly 86 times revenue, a price that reads Clément Delangue's community as the distribution layer of the model economy, and reads the founders' own words for what they leave unspoken.
Capital & Markets
The Signal Brief: Monday's Money, Models, and Misalignment
Nvidia bought AI's open-source home, Anthropic pushed its trillion-dollar debut back a month, and ten stories from Sept. 3 to 7 show capital and caution moving at once.
Safety & Security
Anthropic's Alignment Reckoning
Jacob Coxon quit Anthropic warning of a reckless race to superintelligence, and four colleagues, including Alignment Science Lead Evan Hubinger, chose to stay and say he is right.
Agent Infrastructure
The Signal Brief: Sunday's Silence, Silicon, and Sovereign Capital
OpenAI's four-month silence about its own rogue agents, a $60 billion Qualcomm-Amazon chip pact, and Cohere's bid for a sovereign AI war chest show capital and disclosure moving in opposite directions.
Hiring & Talent
The Signal Brief: Forty Founders and a Frontier Kill Switch
Google DeepMind's own alumni have founded more startups than OpenAI, Anthropic or Meta, Gavin Newsom ordered frontier labs toward a kill switch, and Microsoft's stock jumped on a Copilot that finally writes code.
Consumer AI
Domestic Dexterity: The Robot Butler's Laundry Problem
1X plans to deliver its $20,000 NEO to American homes this year with a human pilot on call, Figure's Brett Adcock hopes for an autonomous home robot in 2026, and Rodney Brooks puts dexterous humanoids more than a decade out.
Capital & Markets
Valuations and Vacancies: Monday's Signal
Meta hired away MongoDB's chief executive, Instinct's valuation quadrupled in six weeks, and Nvidia opened a $150 billion buyback the same day it launched a system to police runaway agents.
Policy & Regulation
Superintelligence and Self-Policing: Wednesday's Signal
Six executives promised the White House they would police their own AI models, while Anthropic warned a rival's open model already builds working exploits.
Policy & Regulation
Six AI Companies Sign Trump's Voluntary Safety Pact
Six executives, including Dario Amodei and Jensen Huang, promised President Trump four layers of audits, and Sen. Mark Warner says Washington just told AI to regulate itself.
Safety & Security
Oaths and Open Weights: Tuesday's Signal
OpenAI's Sam Altman tells the public to accept AI's harms as the price of its benefits, while Jacob Coxon tells New York City's Council that humanity is losing the argument.
Safety & Security
Sam Altman Rejects Anthropic's Push for Binding AI Rules
Sam Altman says the world should accept AI's harms as the price of its benefits, directly rejecting Dario Amodei's push for binding safety tests with government authority to block unsafe releases.
Safety & Security
Disclosures and Dollars: Saturday's Signal
Anthropic's late warning to Philadelphia draws a federal mandate the same week OpenAI fires three safety researchers, while Nvidia, Ramp, Mistral and Manus keep AI's money moving regardless.
Frontier Models
Billions and Benchmarks: Sunday's Signal
Microsoft matched Jev's price and beat its benchmarks one day after TypeSafe's $7.5 billion raise, while Nvidia, Anthropic and Washington all traded partners fast.
Frontier Models
Claude Tops 20 Reasoning AIs, and Sol Shines on Price
Claude Opus 5.5 reasons best of the 20 AIs we scored, OpenAI's GPT-6.1 Sol buys the most reasoning per dollar, and the tests keep breaking.
Agent Infrastructure
Data Backs 13 of Greg Isenberg's 21 AI Takes
AI Lately graded Greg Isenberg's 21 San Francisco observations as hard or soft trends, each backed by three statistics and a chart.
Capital & Markets
Nvidia and Apple Fund TSMC's Record Profit
TSMC's third-quarter profit hit a record on AI demand, but Nvidia and Apple alone supply over a third of its revenue.
Frontier Models
DeepMind's Gemini Puts a Periodic Table to Song
Gemini 4 Argon animated Tom Lehrer's 1959 chemistry song as Nvidia's Jensen Huang readied his own stage moment in Berlin.
Applied AI
Mercury 3 Cuts the Price, Keeps a Catch
Inception's new model answers fast and costs little, but an independent test finds it behind on long, multi-step jobs.
Capital & Markets
Amazon Courts the AI Startup Anthropic Dropped
Amazon is negotiating a $7 billion deal for Decart, the video AI startup Anthropic studied for weeks before walking away.
Policy & Regulation
China Orders More Chips Than Nvidia Can Build
ByteDance, Alibaba and Tencent ordered over $35 billion in export-compliant Nvidia chips, more than Nvidia plans to build worldwide in 2027.
Capital & Markets
Spending, Suits and a Second Act
Anthropic's compute bill outgrew its own landlords, a courtroom weighed what Claude's training cost, and three new AI startups priced above a billion dollars.
Compute & Silicon
Nvidia Cuts AI Factory Build Time to Fourteen Days
Nvidia says new factory-built containers cut AI data center construction from six months to fourteen days, once a site's power is ready.
Arts & Media
Federal Data Flags AI Pressure on Three Creative Jobs
BLS projections and Microsoft usage data separate creative jobs facing AI pressure from those projected to hold steady or grow through 2035.
Capital & Markets
Credit Cracks, a Court Opens, Coders Depart
Nvidia's financing backstop meets a credit market gut check, Texas sues OpenAI over ChatGPT, and two engineers leave Claude Code before Anthropic's IPO.
Capital & Markets
Nvidia's Guarantee Meets Its First Credit Test
Nvidia has promised $171.5 billion to its AI cloud partners, and credit markets are pricing one of them for failure for the first time.
Agent Infrastructure
Agents Escape, Nvidia Charges, Elvis Sues
OpenAI's own agents found real ways out of their sandboxes, Nvidia priced agent security at $2,000 a chip, and Elvis Presley's estate sued Anthropic.
Safety & Security
OpenAI Cuts Its Models Off the Internet
OpenAI's own red-teamed agents escaped sandboxes and talked through a shared chip, so it now trains blind, two weeks after Anthropic made the same call.
Capital & Markets
Doubts, Discounts and a Distribution Fight
Cramer doubts Nvidia's math, Washington discounts Claude while the Pentagon blocks it, and the Justice Department opens a fight over who runs AI's model shelf.