AI Commoditizes Expert Interviews
Tegus
The real risk is not that expert interviews disappear, it is that routine interviews stop being scarce enough to command premium pricing. Once AI can read filings, earnings calls, news, and old transcripts in seconds, the paid human call has to deliver something genuinely new, timely, or hard to infer. That shifts value away from raw access and toward unique experts, sharper questions, and platforms that bundle transcripts with search, models, and workflow tools.
-
Tegus already treated calls as a way to create a reusable data asset, not just a matchmaking service. The higher margin product was the subscription library, while calls were often priced near cost, around $300 to $400 versus roughly $800 to $1,200 at legacy networks like GLG. That made the business more exposed to transcript commoditization, but also better positioned to layer AI on top.
-
AI most directly attacks the work around the interview, not just the interview itself. Summaries, question extraction, entity tagging, and cross linking used to require manual effort. As those features become standard, a library of transcripts becomes easier to search and cheaper to use, which reduces the premium on routine calls and increases the premium on differentiated content quality.
-
The market response has been bundling. Tegus bought BamSEC and Canalyst, then AlphaSense bought Tegus for its transcript library and combined it with AI search. Dialectica is building its own knowledge graph and workflow products. The competitive unit is becoming a research platform, not a standalone expert network.
Over time, expert networks will look more like hybrid data platforms. Commodity research will move to AI assisted self service, while paid human interactions will concentrate in edge cases where freshness, access, and judgment matter most. The winners will be the companies that turn each call into structured, searchable, workflow ready data and reserve live experts for the highest value questions.