The race toward artificial general intelligence (AGI) has become a defining goal of the AI industry, with leading experts offering divergent predictions on when this transformative technology will arrive. AGI represents a hypothetical form of machine intelligence capable of solving human tasks through methods not constrained by its training data—essentially matching or exceeding human cognitive abilities across diverse domains.
Google DeepMind CEO Demis Hassabis, a recent Nobel laureate, predicts AGI will emerge “in the next five to ten years” during his April 20 appearance on 60 Minutes. By 2030, he envisions systems that “really understand everything around you in very nuanced and deep ways” and become embedded in everyday life. However, Hassabis acknowledges current limitations: AI systems still lack curiosity, imagination, and intuition, and cannot yet formulate novel questions or hypotheses independently.
OpenAI CEO Sam Altman takes an even more optimistic stance, suggesting on the Y Combinator podcast that we’re already making major progress toward AGI, with 2025 potentially marking its arrival. His former colleague Miles Brundage, OpenAI’s ex-head of AGI readiness, predicts that within the next few years, AI systems will “basically do anything a person can do remotely on a computer,” including operating interfaces and appearing human-like in video chats.
Anthropic CEO Dario Amodei believes AGI—which he prefers calling “powerful AI”—will manifest by 2026. He describes it as intelligence smarter than Nobel Prize winners across multiple fields, multimodal, independent, fast, and replicable, likening it to “a country of geniuses in a data center.”
AI pioneer Geoffrey Hinton offers a broader timeline of 5 to 20 years for AI surpassing human intelligence, though he admits “without much confidence” given the uncertainty of the times. Meanwhile, AI researcher Andrew Ng takes a more conservative approach, expressing hope for AGI “in our lifetime” but urging skepticism toward companies claiming imminent arrival.
Richard Socher, CEO of You.com, presents two definitions: an economic one where 80% of jobs are automated (3-5 years), and a comprehensive human-like intelligence (10-200 years). Meta’s chief AI scientist Yann LeCun stands as the most skeptical, stating at Davos that AGI “is not around the corner” and will take “years, if not decades.” He emphasizes that AGI won’t arrive as a single event but through gradual progress, rejecting the Hollywood narrative of a sudden breakthrough.
Key Quotes
In the next maybe five to ten years, I think we’ll have systems that are capable of not only solving a important problem or conjecture in science, but coming up with it in the first place.
Google DeepMind CEO Demis Hassabis explained on 60 Minutes that while current AI systems lack curiosity and imagination, the next generation will be capable of formulating novel scientific questions and hypotheses—a crucial step toward true AGI.
There’s a simple economic one, which is 80% of the jobs will be automated with AI, and then we can call it AGI.
You.com CEO Richard Socher offers a pragmatic definition of AGI focused on economic impact rather than purely technical capabilities, predicting this milestone could arrive within 3-5 years and fundamentally reshape the labor market.
The idea somehow which, you know, is popularized by science fiction and Hollywood that, you know, somehow somebody is going to discover the secret, the secret to AGI… That’s just not going to happen. It’s not going to be an event. It’s going to be gradual progress.
Meta’s chief AI scientist Yann LeCun pushes back against the narrative of sudden AGI emergence, arguing on Lex Fridman’s podcast that the technology will develop incrementally over years or decades rather than appearing as a single breakthrough moment.
By 2030, we’ll have a system that really understands everything around you in very nuanced and deep ways and kind of embedded in your everyday life.
Demis Hassabis provides a concrete vision of near-term AGI capabilities, suggesting that within six years, AI systems will achieve contextual understanding sophisticated enough to integrate seamlessly into daily human activities.
Our Take
The stark divergence in AGI timelines among top experts reveals more than just technical disagreement—it exposes fundamental definitional ambiguity about what constitutes artificial general intelligence. Hassabis’s focus on scientific hypothesis generation, Socher’s economic automation threshold, and Amodei’s “country of geniuses” metaphor represent fundamentally different goalposts.
What’s particularly striking is the gap between commercial optimism and scientific caution. CEOs of AI companies (Altman, Amodei) predict imminent AGI, while independent researchers (Ng, LeCun) urge skepticism. This pattern suggests competitive pressures and funding dynamics may be influencing timelines as much as technical assessment.
LeCun’s gradual evolution perspective appears most credible given AI’s historical development pattern. The current limitations Hassabis identifies—lack of curiosity, imagination, and novel reasoning—represent profound architectural challenges unlikely to be solved through scaling alone. True AGI may require paradigm shifts in AI design rather than incremental improvements to existing transformer-based models.
Why This Matters
These AGI predictions from the world’s leading AI experts reveal both the ambitious trajectory and fundamental uncertainties shaping the AI industry’s future. The wide range of timelines—from Altman’s 2025 optimism to LeCun’s decades-long projection—highlights the lack of consensus on what constitutes AGI and how to measure progress toward it.
This matters profoundly for business strategy, workforce planning, and policy development. If AGI arrives within 3-5 years as some predict, organizations face unprecedented disruption requiring immediate preparation. The potential automation of 80% of jobs, as Socher suggests, would fundamentally reshape labor markets and economic structures.
The technical limitations identified by Hassabis—lack of curiosity, imagination, and novel hypothesis generation—reveal critical gaps between current AI capabilities and true general intelligence. This suggests that despite impressive advances in large language models and multimodal systems, fundamental breakthroughs in AI architecture may still be necessary.
For investors, policymakers, and business leaders, these divergent expert opinions underscore the importance of scenario planning rather than betting on a single timeline. The gradual evolution LeCun describes may be more realistic than sudden transformation, requiring sustained investment and adaptive strategies rather than reactive crisis management.
Recommended Reading
For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources:
Recommended Reading
Related Stories
- Sam Altman’s Bold AI Predictions: AGI, Jobs, and the Future by 2025
- Artificial General Intelligence Could Arrive by 2024, According to AI Experts
- Google’s Gemini: A Potential Game-Changer in the AI Race
- The Artificial Intelligence Race: Rivalry Bathing the World in Data
- The AI Hype Cycle: Reality Check and Future Expectations