Will OpenAI publicly declare that they have achieved 'Artificial General Intelligence' (AGI) based on their own internal definition before the end of 2027?

AGI Declared by 2027?

Sam Altman and OpenAI have set AGI as their explicit mission. This prediction asks if they will make a public, formal declaration that their model (e.g., GPT-6 or a reasoning agent) has met their criteria for AGI before the end of 2027.

Moving the Goalposts

The 'No' vote leads because the definition of AGI is notoriously slippery. Even if models achieve superhuman performance in coding or research (which Altman predicts for 2028), declaring 'AGI' carries massive safety, regulatory, and PR liabilities. It is more likely OpenAI will describe their systems as 'autonomous agents' or 'reasoning engines' to avoid the panic and regulation that a formal AGI declaration would trigger in the short term.