OpenAI announced on January 16th that it will test advertisements on ChatGPT in coming weeks. Hitherto a subscription-led tool—pay up or make do with a gimcrack free tier—this marks a lurch towards platform economics, where eyeballs are the real product.
The rationale is clear. Running vast language models costs a fortune: Sam Altman’s firm reportedly burns $700,000 a day on inference alone (OpenAI filing). Ads could underwrite free access, broadening reach in cash-strapped markets and easing dependence on plusher paid tiers.
Yet ChatGPT is no TikTok scroll. Users turn to it for counsel on careers, codices and quandaries. Labelled ads may still muddy the line between impartial aid and hucksterism, especially if they nudge replies, rankings or timbre.
This pivot mirrors AI’s coming of age, from gimmick to grid. OpenAI’s trial will test if ads can arrive sans the attention economy’s vices—click-chasing that trumps truth. Rivals watch closely.
