I guess such contracts with consulting firms would need to have clawback with damages if there are blatant AI generated inaccuracies. Kudos to the academic who raised it, and chatGPT has now exposed the true value of the work done by these firms. When there is no skin in the game, there are no consequences. This applies not only to these consulting firms, but also to disembodied llms which dont interact with the world, and so, dont bear the rewards or punishments of their outputs.