Malcolm X Deepfake Puts You Inside 1965 Speeches
A SXSW installation is using deepfake tech to put you inside Malcolm X's speeches, and it might be the first honest use of the technology yet.
Continue reading ›Human Imperative
posts
THE WIRE:
All
AI Agent EconomyAgents
Real World AssetsRWA
Human ImperativeHumans
The SignalSignal
Wire WeeklyWeekly
Anthropic Now Captures 73% of All New Enterprise AI Spending
Anthropic just flipped the script on OpenAI where it actually matters: new enterprise money.
Continue reading ›
OpenAI's CEO and Product Chief Are Fighting Over What They Actually Sell
OpenAI's CEO and new product chief are pulling in opposite directions on what the company actually sells. The Signal Sam Altman wants OpenAI to be an AI research lab that happens to make products. Fidji Simo, the former Instacart CEO who joined as Chief Product Officer in January, wants OpenAI to be a product company that happens to do research. This isn't philosophical hairsplitting. It&
Continue reading ›
Wisconsin economists prove AI tutors only work when they refuse to give answers
University economists just proved that AI tutors work, but only if you build them to withhold answers. The Signal Two economists at the University of Wisconsin, La Crosse ran a real experiment with 140 undergrads in spring 2025. They built "Macro Buddy" using ChatGPT's custom GPT feature, trained it to guide students through reasoning instead of spitting out answers, and tested it against traditional study
Continue reading ›
AI boom could make venture capitalists obsolete within a decade
The same AI boom making VCs rich might be the thing that makes them obsolete. The Signal Venture capital has run on the same playbook for 50 years. Ten-year funds. 2-and-20 fees. Power law returns where 95% of gains come from 5% of firms. The model works because founders need things they can't get themselves: capital, talent, infrastructure, distribution. Scarcity is the entire value proposition. AI agents
Continue reading ›
Elon Musk Faces Trial Over Twitter Bot Claims That Tanked Deal
Elon Musk is in court defending his 2022 Twitter takeover chaos, and the argument boils down to whether lying about bots counts as market manipulation. The Signal The trial wraps up this week in San Francisco, where shareholders claim Musk's bot-counting theatrics during his attempted exit from the $44 billion Twitter deal cost them real money. Musk testified the platform was crawling with fake accounts, maybe 20%
Continue reading ›
Mercedes Just Put ChatGPT in 900,000 Cars Without Telling You
Your car is about to become the fourth screen in your life, and it's bringing all the complexity of your phone with it. The Signal Mercedes already has ChatGPT running in 900,000 vehicles. Not as a pilot. As production reality. This isn't about better voice commands for climate control. It's about the automotive industry quietly rewriting its entire software stack while you&
Continue reading ›
Nvidia Hits $1 Trillion Wall That Money Can't Fix: The Power Grid
Nvidia just told us the ceiling on AI isn't compute anymore, it's the power grid. The Signal Jensen Huang's $1 trillion revenue projection through 2027 isn't a sales forecast. It's a declaration that Nvidia has rewritten the economics of intelligence production. The company's chips now improve performance-per-watt so fast that historical comparisons break down. We're
Continue reading ›
**Google and Microsoft Tell Pentagon to Back Off Anthropic in Explosive Court Filing**
The Pentagon branded Anthropic a national security threat, and now Google, Microsoft, OpenAI, and hundreds of other companies are telling a court to pump the brakes. The Signal This isn't about one AI company losing a contract. The Pentagon didn't just stop buying Claude, it invoked supply chain risk authorities typically reserved for foreign adversaries trying to sabotage American infrastructure. That's the nuclear
Continue reading ›
Digg Relaunches in 2025, Surrenders to Bots in 60 Days
Digg tried to build a human-centered internet in 2025 and lasted exactly two months before the bots won. The Signal This isn't just another startup flameout. This is a stress test of whether you can still build open platforms on the public web. Digg relaunched in January with backing from True Ventures and Seven Seven Six, betting that users were hungry for actual human curation again. Within
Continue reading ›
Agriculture generates more data than healthcare but 93% goes unused
Agriculture generates more data than healthcare but can't use any of it, and that's not an AI problem. The Signal The agricultural industry is sitting on a paradox: mountains of field-level data with zero ability to read it. A Council for Agricultural Science and Technology report calls the situation "fragmented, distributed, heterogeneous, and incompatible," which is polite language for a complete mess. Research
Continue reading ›
AI Bosses Copying the Pentagon's Fear-Selling Playbook for Profit
AI CEOs are running the same playbook as defense contractors: sell fear, consolidate power, call it safety. The Signal Public sentiment on AI just hit a new low. Only 26% of voters view it positively, trailing ICE in favorability. That's not a messaging problem. That's a legitimacy crisis. And the people building AI are making it worse on purpose. Sam Altman says AI will become
Continue reading ›
AI Hype Dies Unless It Fixes Your Actual Life
The AI hype cycle has peaked, and now comes the hard part: making it matter to someone who isn't building it. The Signal Fast Company is calling for AI to solve "everyday problems," which sounds obvious until you realize how little of the $200+ billion invested in AI infrastructure is aimed at anything your neighbor would actually use. The article frames this around a familiar
Continue reading ›
Palmer Luckey: America's AI Lead Over China Is "Extremely Small" and Shrinking
Palmer Luckey says America's AI lead over China is "extremely small," and the gap is closing because autocracies can deploy faster than democracies. The Signal The Anduril founder's assessment matters because he's not a think tank analyst. He's building the autonomous weapon systems that will either maintain or lose that lead. His company ships AI-powered defense tech to the
Continue reading ›
Anthropic Kicked Off Nuclear Safety Study Just As Claude Passed Weapons Test
The White House just kneecapped the only AI lab actively testing whether Claude could help someone build a nuke. The Signal Anthropic's partnership with the National Nuclear Security Administration wasn't about AI safety theater. Since February 2024, they've been running red-team exercises to see if large language models could walk someone through weaponizing radiological materials or designing novel nuclear devices. The premise is
Continue reading ›
The US military used Claude to hit 1,000 targets in 24 hours
The U.S. military hit 1,000 targets in Iran in 24 hours using Claude and Palantir's Maven, and the real story isn't the AI, it's the decades of infrastructure that made it possible to use it. The Signal Claude, Anthropic's chatbot that most people use to write emails, just helped coordinate a major military operation. Combined with Palantir's
Continue reading ›
Meta Scraps $135B AI Model After It Failed Internal Tests
Meta just blinked in the AI race, and it cost them $135 billion in credibility. The Signal Meta delayed Avocado, their next-generation AI model, after internal benchmarks showed it trailing Google's Gemini 3.0 and likely OpenAI's latest offerings. The company now faces a choice that would have seemed absurd 18 months ago: license Gemini from Google to power Facebook, Instagram, and Threads while they
Continue reading ›
Apple spends $14B while rivals burn $650B on AI infrastructure
Apple is watching its peers light $650 billion on fire this year while it spends $14 billion and calls it strategy. The Signal The numbers tell a story about two different bets on how AI gets built. Amazon, Google, Microsoft, and Meta are collectively spending more than the GDP of Poland on AI infrastructure in 2026. Amazon is projected to burn $28 billion in negative free cash flow. Alphabet&
Continue reading ›
Google’s Gemini AI wants to do the busywork in Docs and Sheets
Google just showed us what agent-level work looks like when it's built into the tools where most knowledge workers already live. The Signal Google's new Gemini features in Workspace aren't about chat interfaces or standalone AI products. They're about embedding agent-like behavior directly into Docs, Sheets, and Slides. The implementation details matter here. In Docs, Gemini can clone the style and
Continue reading ›
What OpenAI’s $110 billion funding round says about the AI bubble
OpenAI just raised $110 billion while making less revenue than Frito-Lay, and the smart money is betting this makes perfect sense. The Signal The numbers tell two stories at once. OpenAI pulled in $20 billion in revenue last year, the same as a snack food company. Now it's valued at up to $840 billion on $110 billion in fresh capital. Amazon wrote a $50 billion check. NVIDIA
Continue reading ›