Fact check: Did AI labs secretly reach AGI?

25%

IsItCap Score

Truth Potential Meter

Not Credible

AI consensusWeak

Grader consensus is weak.

Range 10%–50% (spread Δ40).

The graders diverge. Treat the combined score as uncertain and read the sources carefully.

Read analysis summary

OpenAI Grade

20%

40%

60%

80%

10%

Perplexity Grade

20%

40%

60%

80%

50%

Google Gemini Grade

20%

40%

60%

80%

50%

Analysis Summary

The claim that major AI labs have secretly achieved AGI is false. Mainstream sources, including Reuters, report that AGI remains a future goal without verified evidence of its achievement. Experts predict low probabilities for AGI being reached in the near term, with leading labs publicly stating timelines of 2-5 years for human-level capabilities, indicating an open race rather than secrecy. Disputes arise from anecdotal claims and speculative videos suggesting otherwise, but these lack credible backing. The models diverge sharply — treat this as higher-uncertainty. Gemini comes in highest (50%), while OpenAI is lowest (10%). OpenAI expresses higher confidence than Gemini on this claim. While some sources, including Reddit posts and YouTube videos, suggest insider knowledge of AGI development in secrecy, these claims are not substantiated by verified evidence. The lack of credible sources supporting the assertion of a concealed AGI breakthrough undermines its validity. The speculative nature of these claims does not alter the overall conclusion that major AI labs have not achieved AGI as claimed, thus maintaining the verdict of false.

Jump to source analysis

Source quality

Truth (from sources)1.00 / 10

Source reliability3.00 / 10

Source independence4.00 / 10

Claim checks

Fits established facts2.00 / 10

Logical consistency2.00 / 10

Expert consensus2.00 / 10

Related fact checks

Source Analysis