OpenAI’s incremental internal roadmap and absence of any AGI declaration continue to anchor trader consensus against an announcement before 2027. Leadership has publicly targeted AI research interns by September 2026 and fully automated researchers only by March 2028, while recent large language models such as GPT-5 variants and o3 have advanced agentic reasoning, coding, and tool use without demonstrating the broad, human-level generality across novel domains that would qualify as AGI. Sam Altman has shifted emphasis toward superintelligence timelines around 2028, and Microsoft partnership terms require independent expert verification before any such claim. Near-term model iterations and benchmark results on evaluations like ARC-AGI remain the primary catalysts that could shift sentiment.