Seth Klarman, founder and CEO of the Baupost Group, believes artificial intelligence has the potential to reshape industries in the future—but for now, its role in professional investing is far more modest. Speaking on Columbia Business School’s Value Investing with Legends podcast, Klarman described AI as a practical tool for speeding up certain research and administrative tasks, not as a substitute for investment judgment.
“We’ve started using it as essentially a capable assistant, a summer intern,” Klarman said. “Not someone who knows which stocks to buy, but a way to tabulate data faster.”
At Baupost, AI is being applied to tasks that once required days of work from junior analysts. One example: a colleague asked an AI tool to review 10 years of a company’s annual reports and identify changes from year to year. The AI flagged differences in tone, structure, and emphasis—subtle cues that might reflect shifts in business strategy, operational priorities, or legal concerns. By automating the comparison process, the team gained deeper insights faster and freed up analysts to focus on interpreting the findings.
In another case, AI was tasked with identifying company logos in a specific sector that one of Baupost’s analysts had never encountered before. The project, which might have taken an intern three days, was completed by the AI in about five minutes, giving the team a much quicker grasp of the industry’s competitive landscape.
Klarman has experimented with AI in his own work, though he remains critical of its creative output. Before a speaking engagement with a prominent business leader, he asked a chatbot to suggest questions to ask the person. The results were disappointing. “What came back was useless,” he said. “I like to ask things they haven’t thought about before.”
Baupost’s use of AI mirrors a broader trend across the financial services industry. Morgan Stanley and Bank of America are training employees to integrate AI into daily workflows, aiming to improve productivity and streamline operations. Research from consulting firm ThoughtLinks projects that by 2030, 44% of a bank’s tasks—including nearly one-third of sales and trading activities—could be “redefined” by AI.
For RIAs and wealth managers, Klarman’s comments illustrate a balanced approach to AI adoption: leverage its speed and scalability for repetitive or data-heavy work, but keep core analytical, strategic, and creative thinking firmly in human hands. The near-term value lies in using AI to enhance the advisory process—aggregating and summarizing information, surfacing patterns, and handling routine tasks—so advisors can devote more time to client strategy and relationship management.
Klarman also voiced concerns about AI’s potential downsides, particularly its impact on creativity and critical thinking. A recent MIT study supports this view, showing that reliance on AI can reduce memory retention and brain activity. For investment professionals, that could translate into a diminished ability to connect dots, identify original insights, or think several steps ahead.
“I think that if we use AI the wrong way, we’ll solve the problem without having applied our brains,” Klarman said. He likened overreliance on AI to skipping ahead to the last page of a novel without reading the story. “I think the right order is, ‘Here’s what I think,’ and then I can improve my thinking.”
For wealth advisors, this is a critical takeaway. AI can be a powerful accelerator, but if it becomes a crutch, it risks weakening the very skills that define effective advisory work—deep analysis, independent judgment, and the ability to navigate complex, evolving situations.
The most effective integration of AI in advisory practices will likely mirror Baupost’s current approach:
-
Targeted use cases – Apply AI to tasks like document analysis, industry mapping, and information summarization.
-
Analyst-first workflow – Let human professionals form hypotheses and strategies first, then use AI to test or expand them.
-
Productivity over prediction – Use AI to save time and broaden coverage, not to dictate investment decisions.
-
Continuous evaluation – Regularly assess whether AI tools are improving outcomes or simply adding noise.
Firms that start experimenting with AI now—without overcommitting—will be best positioned to scale its use as capabilities evolve. Early adoption allows teams to refine internal processes, set usage guidelines, and develop a clear sense of when AI adds value and when it doesn’t.
The Baupost example also offers a cultural insight: integrating AI effectively requires maintaining a mindset of curiosity and skepticism. The goal isn’t to hand over the thinking to the machine, but to let the machine handle the heavy lifting so advisors can think more, not less.
In practice, that could mean assigning AI to handle:
-
Comparative analysis of SEC filings and other regulatory documents.
-
Rapid visual or brand recognition projects to assess competitive landscapes.
-
Initial data aggregation for market trend monitoring.
-
Drafting outlines for research reports, later refined by human analysts.
When used in this way, AI becomes an enabler—boosting efficiency while preserving the intellectual rigor that clients expect from fiduciary advisors.
Klarman’s perspective should resonate with RIAs navigating their own AI strategies. The technology is promising, but it’s not a magic bullet for investment success. Its true power lies in extending an advisor’s reach, sharpening their focus, and allowing more time for client engagement and complex decision-making.
In the end, the winning formula may be what Klarman already practices: keep the thinking human, keep the process disciplined, and let AI work quietly in the background—doing the work no one will miss, and freeing up advisors to do the work that truly matters.
