Ken Griffin once dismissed artificial intelligence as garbage. That was years ago at Davos. Now the founder of Citadel sees agentic systems compressing months of elite research into mere hours. And the implications stretch far beyond trading floors.
Citadel manages roughly $68 billion with a track record that stands out even among elite funds. Average annual returns hit 19.2 percent after fees since its 1990 launch. The firm employs 260 PhDs. They sift through 100 petabytes of data for daily decisions. High-frequency trading? Citadel helped popularize the term. The Motley Fool captured Griffin’s evolving stance in a recent profile.
But skepticism lingers. At the JPMorgan Robin Hood Investors Conference last October, Griffin stated that while generative AI boosts productivity, it “just falls short” for uncovering alpha. Hedgeweek reported his comments. He founded Citadel in 1990. The firm now oversees about $69 billion. No technology has replaced the deep research his teams conduct.
Short sentences. Direct observations. Griffin’s recent appearances reveal a more nuanced position. He recounted asking business leaders how AI transformed their operations. “I couldn’t help myself. I’m like, ‘Let’s go around the table and share stories about how AI is transforming your business,’” he said, per the Yahoo Finance article drawn from The Motley Fool. Responses poured in. Four or five “incredible stories” centered on productivity gains. None, he concluded after probing, actually involved AI.
Data optimization. Digitization. Basic tools often wore the AI label. “The nuance between AI and technology writ large gets a little bit lost,” Griffin observed. He added, “There is a technological revolution happening, of which AI is a component of the story, but it’s just a piece.” Investors should stay wary of corporate claims. Nvidia. Quantum players like Rigetti Computing or D-Wave Quantum. Exposure to genuine AI demands scrutiny.
Productivity leaps inside Citadel itself forced a reckoning.
Academia pumps out finance papers constantly. Citadel’s young analysts and PhDs once spent six to eight weeks reproducing each one. They tested hypotheses. They checked persistence out of sample. Buybacks and stock outperformance, for instance. A few ideas emerged yearly. Some proved valuable. Then an internal team built an agentic AI system. It read the paper. Reproduced the work. Verified results. Ran out-of-sample tests. Average time? Two to three hours.
“This is not just a white collar job. This is a master’s or PhD level job,” Griffin explained in a Stanford Graduate School of Business conversation detailed by Stanford GSB. He went home one Friday “fairly depressed” after witnessing the speed. “You could just see how this was going to have such a dramatic impact on society,” he told the audience. NDTV covered the remarks from that June 2026 event.
But headcount at Citadel did not drop. Talent remains scarce. Problems abound. “I will take every single productivity gain I can get because with the talent of people we have, we just have more to go after,” Griffin noted. Machine learning entered trading fast after Google open-sourced TensorFlow. Within 10 days, it influenced nearly one in four U.S. equity trades. Griffin shared that history to frame generative AI as another incremental step.
Limitations persist for long-horizon investing. Models train on past data. Markets demand forecasts of tomorrow. Short-term signals? They excel. Longer views? They fray. Self-driving cars illustrate the point. Snow disrupts their learned patterns. Investing faces constant novelty. So Griffin views AI as a productivity enhancer inside his firm. Not a replacement for judgment.
Yet the upside excites him. Agentic AI could erase corporate moats. Lower startup barriers. Spark a “golden age of entrepreneurial activity.” He made that case in a July 2026 Goldman Sachs Exchanges podcast. Goldman Sachs published the discussion. Compute costs create new barriers for some players. Data centers must expand domestically. Geopolitical risks with China and Taiwan chip production add urgency.
Call centers face extinction in five to 10 years. Documentation teams shrink dramatically. One multinational friend expects staff to drop from 8,000 to 1,000 within three years. High-paying roles vanish. Retraining becomes essential. “With machine learning is going to come with a cost to society, a cost that we need to understand how do we help these people land on their feet so we don’t end up with a backlash against AI and machine learning,” Griffin warned in the Stanford session.
His funds performed well amid 2025 volatility. The tactical trading vehicle gained 6.1 percent in the first half. The main Wellington multistrategy fund rose 2.5 percent. CNBC noted the results. Griffin increased stakes in names like Microsoft last year. But his AI views shape broader capital allocation.
Society must prepare. Jobs evolve. Some disappear. Others emerge. Griffin sees AI enabling marketing at individual scale. One pet insurance startup used social data to target new dog owners instantly. Sold for substantial gains. Radical applications will define winners over the next decade or two.
At Citadel, the 260 PhDs keep working. AI accelerates their output. No reduction in staff. More problems attacked. The firm’s analytical edge sharpens. Returns compound. Yet Griffin cautions against hype. Many claims mislabel old tools. True breakthroughs remain selective.
So the hedge fund star changed his tune. From garbage to guarded optimism. Productivity soars. Alpha stays hard won. And the societal ledger? Still being tallied. Investors and executives alike should listen closely. The next phase arrives faster than expected.
WebProNews is an iEntry Publication