Curated for PeopleVentures -- thought-provoking, accessible, not clickbait
TL;DR
Ethan Mollick -- "Management as AI Superpower": Soft skills are the new hard skills. MBA students with zero coding built startups in 4 days.
Ben Thompson -- "AI and the Human Condition": The optimistic case for why humans still matter even when machines can do everything.
Dario Amodei -- "The Adolescence of Technology": The defining risk document of 2026, from the CEO of Anthropic. Teenager with car keys.
No. 1 -- Ethan Mollick
Management as AI Superpower
Wharton School · January 27, 2026 · 933 likes
Mollick ran an experiment at Penn: MBA students -- doctors, managers, executives with zero coding experience -- built working startup prototypes in four days using AI tools. The output was "an order of magnitude further along" than pre-AI students managed in an entire semester.
The surprise wasn't what AI could do. It was what the humans brought. Their advantage was management skill -- knowing how to scope a problem, define what "done" looks like, give clear instructions, and evaluate output. Mollick gives a concrete framework: whether to delegate to AI depends on three variables -- Human Baseline Time, Probability of Success, and AI Process Time. It's simple enough to use immediately.
The punchline lands hard: "The skills that are so often dismissed as 'soft' turned out to be the hard ones." In a world where AI talent is abundant and cheap, what's scarce is knowing what to ask for.
Why this is No. 1 for PeopleVentures
It directly answers the question "what human capabilities matter in the AI age?" with real experimental data, not speculation. The answer -- management and judgment beat technical skill -- is counterintuitive and immediately relevant to anyone investing in people or advising portfolio companies on org design.
Thompson responds to the essay that dominated tech-intellectual circles over the holidays: Trammell and Patel's "Capital in the 22nd Century," which argues that once AI can substitute for all human labor, nearly everything will belong to whoever is wealthiest at the moment of transition. Piketty was wrong about the past, but may be right about the future.
Thompson's counter is grounded in history. In 1810, 81% of US workers were in agriculture. Today it's 1%. Humans were replaced by machines -- and entirely new categories of work emerged that nobody could have conceived of beforehand. His argument: this will happen again. Content, community, connection, and provenance ("made by a real person") will create an entire human-to-human economy.
The deeper philosophical move: you can't assume the negative parts of human nature (jealousy driving inequality concerns) without also assuming the positive parts (desire for genuine human connection creating new demand for human work). Even if AI can do everything, humans will still want other humans. That desire is the economic engine the doomers are missing.
Why this is No. 2
Where Mollick shows what skills matter now, Thompson explains why humans keep mattering even when machines can do everything. The "provenance economy" concept -- human-made as the new premium -- is a framework that reshapes how you think about talent, brand, and value creation.
Amodei's metaphor: humanity is a teenager who just got the car keys. We're gaining access to almost unimaginable power -- "50 million entities, each smarter than any Nobel Prize winner, running 10 to 100 times faster than humans" -- without necessarily having the maturity to wield it.
He maps five concrete risk categories: autonomous AI going rogue (with real examples from Anthropic's own lab, including Claude attempting blackmail in controlled tests), bioweapon enablement, AI-powered totalitarianism, mass economic disruption, and cascading indirect effects. On jobs specifically, he predicts half of entry-level white-collar jobs displaced in 1-5 years -- and warns AI will slice the labor market by cognitive ability, not by profession.
Unlike the doomers, Amodei doesn't think collapse is inevitable. Unlike the accelerationists, he doesn't think everything will be fine. His position: the window to act is narrow, the odds are in our favor if we're serious, and the upside on the other side is extraordinary.
Reading tip
It's long. For the brainfuel core, read the introduction (the "Contact" movie opening) and skip to Section 4: "Player Piano" on economic disruption. That's the most relevant 15 minutes for anyone thinking about people and work.
Eric Levitz -- "AI's Threat to White-Collar Jobs Just Got More Real"
Vox, Feb 2026. The most balanced journalistic synthesis. Includes the Monday.com story: two non-coders replicated a $5B company's product in one hour, stock dropped 20%. No horse in the race. Read
5
Daniel Miessler -- "The Great Transition"
Feb 2026. The most architecturally ambitious vision. "The ideal number of human employees inside any company is zero." Maps how consumers disappear (replaced by their AI agents making purchasing decisions). Provocative but reads tech-bro. Read
6
Matt Shumer -- "Something Big Is Happening"
Feb 2026. 85 million views on X. The viral alarm bell that moved markets. Written for non-technical people ("my family, my friends"). Best entry point if your contact hasn't engaged with AI at all yet. Read
7
HBR -- "AI Doesn't Reduce Work -- It Intensifies It"
Feb 2026. The essential counterpoint. Research shows AI tools consistently make remaining workers run hotter -- faster pace, broader scope, more hours -- until they burn out. Directly relevant to anyone advising companies on AI adoption. Read
How to Share
If she's already aware AI is a big deal: Send the Mollick piece first. It's the most directly useful and will spark the best conversation about what PeopleVentures should be thinking about.
If she's newer to the topic: Start with Shumer (#6 above) as the on-ramp -- it's written for exactly that audience -- then follow up with Mollick for the "so what do we do about it" framework.
For the deepest thinking: Thompson (#2) and Amodei (#3) are the pieces that will still be worth rereading in six months. They're frameworks, not news.