Citadel CEO Ken Griffin Admits AI Depression, Reverses Stance in Six Months

Ken Griffin has reversed his stance on AI within six months. On May 16, financial media published his remarks from a Stanford leadership forum, where the Citadel founder admitted to students that AI has improved "more than an order of magnitude in the last nine months" — and then added a headline-grabbing confession:

"I went home one Friday actually fairly depressed by this."

What depressed the Citadel boss was not AI taking his own job, but the jobs of his employees.

The specific scenario that broke him

Griffin detailed the situation on stage. Citadel employs researchers with master's and PhD degrees in finance to handle complex analysis reports that normally take weeks to months. Now, AI agents complete the same work in hours to days.

"It has been really interesting to watch — to be blunt — work that we would usually do with people with master's and PhDs in finance over the course of weeks or months being done by AI agents over the course of hours or days."
"These are not mid-tier white-collar jobs. These are extraordinarily high-skilled jobs being — and I'm going to pick a word — being automated by agentic AI."

Notice his deliberate choice of the word "automated." As the employer of a hedge fund, he publicly admitted for the first time that his most expensive human capital is being automated. What depressed him was not the technology itself, but the realization that "you could just see how this was going to have such a dramatic impact on society."

Six months ago, he had a completely different tone

Looking back at the timeline, the reversal is stark:

  • October 2025: Griffin stated that generative AI had yet to prove it could generate alpha for hedge funds.
  • January 22, 2026 (Davos): He bluntly declared "it's all garbage," calling AI investment a bubble.
  • December 2025: Internally, Citadel had quietly launched an AI assistant for equity researchers to automatically digest regulatory filings, earnings call transcripts, broker reports, and internal strategies — something not mentioned at Davos.
  • May 2026: Public reversal.

A detail often overlooked: while the hedge fund godfather called AI "garbage" at Davos, his own company had already embedded AI into its core research workflow. He criticized "AI can generate alpha," not "AI can save labor."

Four months later, Griffin acknowledged a step change in product capability:

"In the last few months, there has been a step change function in the productivity of the AI toolkit."

He also shared a detail — a colleague used AI to write a commodities report that initially seemed insightful, but deeper analysis revealed "garbage" inside. However, this time he used the example not to dismiss AI, but to illustrate that AI's current weakness is factual inaccuracy, not overall incompetence.

Why this matters

In the hedge fund world, a reversal of this magnitude from Griffin carries signal significance.

First, he has been the most public AI skeptic in finance. His Davos "garbage" comment was cited all winter. Now he retracts it, removing a key pillar from the narrative that "AI is useless in high-skill industries like finance."

Second, his point of depression is precise. Over the past six months, discussions about AI replacing white-collar workers have focused on customer service, junior clerks, and coders. Griffin directly targets finance PhDs — the most expensive brainpower in the industry at firms like Citadel. If even these people are vulnerable, the story for lower-tier jobs is clear.

Third, he did not mention layoffs. Note his tone: he said "work is being done by AI," not "I plan to fire these researchers." An employer still hiring people publicly admitting that their work is being automated is rare, as it is detrimental to internal morale. Griffin chose to say it, indicating he believes the issue outweighs organizational costs.

Finally, the Stanford forum video is publicly available on YouTube. If you work in finance or follow this space, it is worth watching — the transcript does not capture his pauses and word choices.

Sources: Citadel's Ken Griffin Recasts AI Risk Inside High Finance: "Fairly Depressed" (The Deep Dive); AI Is Now Doing The Work A PhD Does In Months In Just Days: Citadel CEO Ken Griffin (OfficeChai); CocoLoop; Ken Griffin Changes Tone on AI, Calls It Real (Let's Data Science)