Random predictions, loosely held Anthropic IPO marks the high water mark of this cycle, the cherry on top. A natural and convenient point to exit all public equities for funds up many multiples in the last few years OpenAI has to go out as well, private markets tapped out for their needs, but reduces valuation as public markets reprice Anthropic. Generally think they trade very similarly to each other once public. The commodity saxophones get louder. If they do end up becoming the utilities of intelligence, does it mean they have limited price appreciation and end up paying a dividend?? IPO window slams shut for all but a select few. Databricks goes public but Stripe continues to delay. They never go public and pioneer a new kind of private company structure because they like to do out of the box things The quarterly reporting cadence of public markets prove unfriendly to the frontier labs and they wish they could've stayed private. Wall Street struggles to digest their business model and rate of spending. Constant model releases fall out deaf ears as people tune out marginal improvements until people are caught off guard by AGI. Market as a whole cools for at least 18 months as everyone waits for "the next breakthrough" Labs further shift focus from frontier intelligence to moneymaking endeavors - vertical and industry focused initiatives in biotech, applied AI deployments, F500 contracts Everything public in the AI ecosystem trades downstream of the labs and acts as levered beta, any weakness in lab numbers is punished severely across the board. Private companies remain insulated from this to a certain extent but the exuberance we see there is significantly dampened as investors now have a public benchmark to work backwards from. NVIDIA buys up all the neolabs for acquihires and backstops the neoclouds but declining stock price due to macro conditions curtails their enthusiasm after a while, they can only cover so much. If they haven't invested in you, you don't spend heavily on GPUs or don't have a personal connection, you're on the outside looking in Surviving SaaS dinosaurs that have good economics are able to get their M&A machines going again to pick up the scraps. Many cutting edge AI startups turn out to have been public R&D projects Using AI to solve real world problems becomes "higher status" than working at a frontier lab for most roles. People become fixated with the idea of real world impact Robotics IPOs are for next cycle, this time around the exits will mainly come via M&A to existing industrial giants who need to catch up