Scott Stevenson, co-founder and CEO of legal AI startup Spellbook, sparked a heated debate in the tech community last month by publicly calling out what he described as a "huge scam" among AI startups. In a post on X, Stevenson accused companies of using "dishonest metrics" to inflate their revenue figures, claiming that major venture capital firms are complicit in supporting this practice and misleading journalists for publicity. His post resonated widely, generating over 200 reshares and comments from prominent investors, numerous founders, and triggering media coverage. Jack Newton, co-founder and CEO of legal startup Clio, told TechCrunch that Stevenson did an excellent job exposing problematic behavior and that his post raised much-needed awareness about the issue, particularly referencing an explanatory post from Y Combinator's Garry Tan about proper revenue reporting. TechCrunch investigated these claims by speaking with over a dozen founders, investors, and startup finance professionals. Sources, many of whom requested anonymity, confirmed that manipulating annual recurring revenue numbers in public announcements is indeed widespread, with some investors reportedly aware of the exaggerations. One investor acknowledged that when competing startups in a category report inflated figures, it creates pressure for others to follow suit to remain competitive. This suggests the practice has become somewhat normalized within the ecosystem, even among parties who recognize its dishonesty. The primary method of obfuscation involves companies reporting contracted or committed ARR (CARR) and simply labeling it as ARR, despite the two metrics measuring different things. Annual recurring revenue was originally designed during the cloud computing era to represent total sales of products where payments are spread over time, with the metric reflecting the value of active customer contracts. However, GAAP accounting standards focus on historical, already-collected revenue rather than future projections, meaning ARR figures are never formally audited or verified by accountants. This lack of formal oversight creates room for startups to present optimistic projections as established revenue, particularly attractive in a market where investors are eager to fund promising AI companies.