Elon Musk has backed a viral social media post forecasting that leading US artificial intelligence labs will achieve artificial general intelligence (AGI) by 2027. His single-word response deemed the prediction accurate, pulling in 1.8 million views within roughly three hours.
The post originated from X user Dr Singularity, who suggested that Anthropic’s Claude Opus 5.5 already feels close to AGI before asserting that Google, OpenAI, SpaceXAI, and other entities will soon reach the milestone.
Musk’s Nod Lands in a Month Crowded With AGI Claims
Earlier in the month, OpenAI President Greg Brockman characterized the September 3 debut of GPT-6 Astra as a “generational leap.”
“Welcome to the AGI era,” he stated during the briefing.
Nvidia CEO Jensen Huang also asserted on X that AGI has arrived, and Musk has previously made similar assertions regarding his own research lab.
On September 14, when a user inquired how close Grok 4.8 would get to AGI, Musk responded on X that AGI will materialize with Grok 5—an exchange previously covered by BeInCrypto.
The latest thread connects this competitive push directly to power availability.
“This prediction is based on the amount of compute they already have + additional compute being added all the time….It looks like few GW of compute are enough for strong AGI, now imagine what we will able to do with 10-20GW of compute which are rapidly coming online,” Dr Singularity wrote.
Traders Keep Their Distance From 2027
Financial prediction markets have not aligned with Musk’s outlook. Traders on Kalshi estimated a 44.3% probability that OpenAI will announce AGI prior to 2028.
Skepticism deepens regarding a timeline within the current year. Polymarket participants assigned OpenAI an 18% likelihood of declaring AGI before 2027, whereas Kalshi placed that figure at 14.8%.
The sentiment is echoed by figures associated with the benchmark OpenAI highlighted upon release. Mike Knoop, co-founder of Zapier and the ARC Prize, described Astra’s performance as a major advance while noting that definitive proof of AGI remains absent.
Warnings Take Aim at the Speed of the Race
At the same time, organizations accelerating toward AGI are among those sounding alarms about its development. In a September 6 essay, OpenAI chief scientist Jakub Pachocki argued that no institution has adequately mastered alignment and safety monitoring.
“I am concerned no one is prepared for the consequences of a continued rapid rise in machine intelligence,” he stated.
Jerome Glenn, who leads the AGI Panel of the UN Council of Presidents of the General Assembly, cautioned that if governments fail to properly oversee the transition to AGI, humanity could forfeit control to machine intelligence.
“AGI could enable some states and malicious non-state actors to build chemical, biological, radiological, and nuclear weapons. Moreover, large AGI-controlled swarms of lethal autonomous weapons could themselves constitute a new category of WMDs,” he warned.
Other experts question whether the current trajectory will culminate in AGI at all. Turing Award recipient Yann LeCun, co-founder of the AI startup AMI Labs, argued via X on September 20 that autoregressive large language models on their own will fail to achieve human-equivalent AI.
He contended that if such an outcome were possible via current methods, consumers would already have access to domestic robots and fully autonomous Level-4 or Level-5 vehicles.
Subscribe to our YouTube channel to watch leaders and journalists provide expert insights