Morning Edition

The Futurist

Tuesday, November 17, 2026
AI & Technology Markets & Crypto Ideas Worth Keeping
S&P 5007,561.67▼ 0.40%
Dow Jones52,347.55▲ 0.10%
Nasdaq25,136.27▼ 0.90%
VIX20.07▲ 4.10%
BITCOIN$64,598.00▼ 1.29%
RIPPLE$0.95▼ 1.08%
ETHEREUM$1,807.80▼ 1.10%
DOGECOIN$0.06▼ 0.47%
01

AI & Technology

404 MEDIA

How an AI Content Mill Put a Real Musician Through Copyright Hell

A company founded last year is claiming copyright ownership over a musician's five-year-old recording. The artist can't reach a human being to dispute it — only automated systems that loop back to the same dead end.

ALL-IN PODCAST

Jensen OUT as Nvidia CEO: "Spy" Scandal, Export Restrictions, Market Impact

Jensen Huang is out at Nvidia amid allegations of leaking classified memos, sending shockwaves through the AI trade. The besties dig into who wanted him gone, what it means for "The Bank of Nvidia," and China's strategic position as the CEO search begins.

ALL-IN PODCAST

Anthropic S-1 Shocker, Greatest Business Ever? Google's $500B Wipeout

Anthropic's S-1 is out and the numbers are turning heads — the besties debate whether it's the greatest business structure they've ever seen. Meanwhile Google shed $500B in market cap in a single day, its worst on record.

A16Z

Why Frontier Models Still Can't Parse a PDF

Reducto CEO Adit Abraham makes the case that the document layer — PDFs, scans, spreadsheets — is the quiet blocker preventing AI agents from doing real work in healthcare, legal, and finance. Solving parsing isn't glamorous, but it may be foundational.

A16Z

Cybersecurity in the Agentic Era

When AI agents can find and exploit vulnerabilities autonomously, the assumptions underlying traditional security collapse. Neo's Nick Warner and Cotool's Max Pollard break down why guardrails built for humans and malware simply don't translate.

SIMON WILLISON

Datasette 1.0a43: Claude Fable 5.2 Pushing Toward 1.0

Simon Willison is leaning on Claude Fable 5.2 to clear the Datasette 1.0 milestone, now 53% complete. A quiet signal that frontier models are becoming genuine co-maintainers on serious open-source projects.

02

X / Twitter Signal

Mollick calls out a striking market paradox: equities are pricing in business-as-usual futures even as CEOs, AI researchers, and government officials openly describe a world being fundamentally restructured by AI. Markets are either rational or asleep — and right now the VIX ticking up to 20 suggests the tension is building.

More than eight months after Fable and Mythos dropped, no open-source model has reached that capability tier. Mollick flags this as contrary to expectations — the open ecosystem that closed the gap on GPT-4 hasn't replicated the trick at the frontier. The implications for who controls the most capable AI are significant.

Google DeepMind's Alex Imas drops a cold economic truth: price competition can keep AI lab revenue flat even as token usage explodes. More usage doesn't automatically mean more money if everyone is racing to cut prices. This is the core tension in every AI bull thesis right now.

AI can now reconstruct what you're looking at from a brain scan alone. Cheung notes we're roughly two papers away from recording dreams. This isn't a parlor trick — it's the opening chapter of a cognitive privacy crisis that no one has seriously regulated yet.

03

The Thread

The Jensen Huang exit is the week's gravitational event. Whether it's espionage, internal politics, or export-control pressure from Washington, losing the face of the AI hardware boom mid-cycle is not a minor personnel change. Nvidia isn't just a chipmaker anymore — it's the financial plumbing of the entire AI buildout. The market is nervous for good reason, with Nasdaq down nearly 1% and VIX climbing.

"Markets are basically assuming the future will look like the past even as many CEOs, AI observers, and government officials openly say otherwise." — Ethan Mollick
The deeper story this week is about automated systems eating human accountability. A musician can't dispute a fraudulent copyright claim because there's no human to call. Cybersecurity teams can't defend against AI agents using tools built to stop humans. PDF parsers still choke on the documents that run trillion-dollar industries. In each case, the automation moved faster than the accountability layer. That's the actual AI risk story — not rogue superintelligence, but systems that work exactly as designed while creating harms no one can easily reverse.

Meanwhile the open-source model gap quietly widens. Eight months past Fable with no open challenger is a long time in this industry. If the frontier stays closed, every downstream application — the copyright bots, the mortgage agents, the security tools — runs on models controlled by a handful of companies currently going through IPO roadshows and leadership crises simultaneously. The Anthropic S-1 lands in that context: a "doomer" lab raising billions to build the thing it warns about, now asking public markets to price the paradox.