🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

The Toronto Trio (Agrawal, Gans, Goldfarb) does it again. Their new paper argues AI's effect on inequality follows a U-shaped curve: First, AI substitutes for skill, decreasing inequality. Then, it complements judgment, increasing it. l.sslevine.com/o4TA

Beyond “human in the loop”: Reciprocal Human-Machine Learning means AI and experts learn from each other—boosting both models and minds. 🤖🧠 MMG Science INFORMS l.sslevine.com/eAvo

Convinced I was the bug after a 4-hour fight with my code. I fed it to Gemini. It methodically ruled everything out... then gave up, blaming the other app. Forget AI hallucinating. The future is AI validating. A tool that knows when to stop is one you can trust.

The endless AI "does it think?" debate became a distraction. Billiard players don't solve physics equations; they just act as if they do. Let's apply the same logic: what matters is the quality of the result, not the internal process. Judge the shot, not the player.

For anyone who has ever questioned the value of a corporate offsite, you are not alone. But true value of a corporate offsite isn't the team bonding; it's the broadcast signal. Brilliant smj paper by @MadelineKneeland & @AdamKleinbaum l.sslevine.com/5ecm

Five charged in the killing of U.C. Berkeley Haas professor, including his ex-wife l.sslevine.com/RYFL

The Toronto Trio (Agrawal, Gans, Goldfarb) does it again. Their new paper argues AI's effect on inequality follows a U-shaped curve: First, AI substitutes for skill, decreasing inequality. Then, it complements judgment, increasing it. l.sslevine.com/o4TA

(Good ideas can take a decade to be accepted. This one was worth the wait.) Congrats to Organization Science on their new research policy. It echoes the path we proposed in MOR back in 2016: Embrace transparent imperfection over feigned certainty. l.sslevine.com/E48g

My original brain has terrible recall. So I built a new one. I vectorized my entire academic library—9,000+ papers—to create Levine Lit. Now I can instantly ask my own reading history the important questions. It's not replacing me yet, though. Still can't forage for cheese cubes.

Stunning nature paper from Marcel Binz et al. introduces Centaur: a single foundation model to simulate human cognition. The shift is profound: We're moving from building computational tools to building a computational theorist. l.sslevine.com/4Ogb

Tenure protects academic freedom—but data shows it may also trade impact for novelty. A massive study of 12,000+ careers finds post-tenure work is more novel but cited less. A core trade-off in academia. l.sslevine.com/UNFq

In 2023, ChatGPT stunned us with perfect mimicry of human language. Now, humans begin to speak like ChatGPT, new paper reveals. The mimic is now the model. A circle closes. l.sslevine.com/XL1y

We often treat symptoms, not causes. A landmark nature study finds only ~25% of the immigrant pay gap is unequal pay for equal work. The real issue? Unequal access to high-paying jobs (~75%). A vital insight for policymakers. l.sslevine.com/8yfx

New paper shows 1890s managers hid good news, revealing a lack of “strategic intelligence.” But signs of learning suggest these skills are teachable. For behavioral strategy, finance, & econ. @ThomasBourveau @MatthiasBreuer @RobertStoumbos @RFS_Editors l.sslevine.com/5ecm

Convinced I was the bug after a 4-hour fight with my code. I fed it to Gemini. It methodically ruled everything out... then gave up, blaming the other app. Forget AI hallucinating. The future is AI validating. A tool that knows when to stop is one you can trust.

The endless AI "does it think?" debate became a distraction. Billiard players don't solve physics equations; they just act as if they do. Let's apply the same logic: what matters is the quality of the result, not the internal process. Judge the shot, not the player.

The Toronto Trio (Agrawal, Gans, Goldfarb) does it again. Their new paper argues AI's effect on inequality follows a U-shaped curve: First, AI substitutes for skill, decreasing inequality. Then, it complements judgment, increasing it. l.sslevine.com/o4TA

(Good ideas can take a decade to be accepted. This one was worth the wait.) Congrats to Organization Science on their new research policy. It echoes the path we proposed in MOR back in 2016: Embrace transparent imperfection over feigned certainty. l.sslevine.com/E48g

Stunning nature paper from Marcel Binz et al. introduces Centaur: a single foundation model to simulate human cognition. The shift is profound: We're moving from building computational tools to building a computational theorist. l.sslevine.com/4Ogb

Tenure is meant to protect academic freedom—but it may come at a cost. A new study of 12,000+ faculty shows that after tenure, researchers pursue riskier, more novel work... that gets cited less. A core trade-off in academia: impact vs. novelty. l.sslevine.com/UNFq