Visualizing the Shift: How Expert AGI Timelines Evolved from 2023 to 2026
Coverage of lessw-blog
A new visualization highlights the shifting predictions of individual AI researchers, offering a granular look at AGI timelines and why aggregate forecasts might be missing the full picture.
In a recent post, lessw-blog discusses a compelling new visualization that tracks how individual expert forecasts regarding Artificial General Intelligence (AGI) timelines have evolved between 2023 and 2026. As the artificial intelligence landscape accelerates, pinpointing when AGI might be achieved has become a central focus for researchers, policymakers, and investors.
Understanding these timelines is critical. The perceived arrival date of AGI directly influences resource allocation, AI safety prioritization, and global regulatory frameworks. Traditionally, the AI community has relied on aggregate forecasts to gauge industry sentiment. While these aggregates provide a useful macro-level view, they often obscure the nuanced, and sometimes dramatic, shifts in individual expert beliefs. A consensus median might remain stable even if half the experts drastically shorten their timelines and the other half extend them. lessw-blog's post explores these dynamics by zooming in on the individual forecaster level.
The author presents a visualization specifically designed to capture net updates in AGI forecasts, offering a complementary perspective to existing aggregate models. By focusing on prominent AI Futures researchers-including Daniel Kokotajlo, Eli Lifland, and Nikola Jurkovic-the analysis reveals how specific experts have updated their timelines. Interestingly, the data shows fluctuations where timelines were initially pushed back later into 2025, only to be revised sooner into 2026, highlighting the volatile nature of predicting exponential technological growth.
To maintain rigor, the visualization employs strict inclusion criteria. Forecasters must have made at least two predictions with specific dates, provided confidence intervals, and utilized a comparable definition of AGI. For the purposes of this analysis, the working definition of AGI is established as the point where most purely cognitive labor is automatable at better quality and speed. This specific threshold moves the goalpost from abstract benchmarks to tangible economic impacts, making the timeline forecasts highly relevant for workforce planning. The author notes a major caveat: definitions of AGI vary significantly among forecasters, and sometimes even shift for the same individual over time. This variability adds a layer of complexity to interpreting the data, yet the visualization successfully captures the broader trend of expert sentiment recalibration.
This granular view is a valuable signal for the AI and machine learning community. Tracking specific researchers' updates offers deeper insight into the perceived acceleration or deceleration of AGI development than broad surveys alone. For those involved in strategic planning or AI risk assessment, understanding the trajectory of these expert opinions is indispensable.
To explore the visualization and understand the specific methodologies behind these shifting predictions, read the full post.
Key Takeaways
- Individual forecast changes provide a more accurate reflection of shifting expert sentiment than aggregate timelines, which can obscure significant personal updates.
- The visualization tracks prominent AI researchers, revealing volatile timeline updates that fluctuate between 2025 and 2026.
- Inclusion in the dataset requires strict criteria, including multiple forecasts with specific dates and confidence intervals.
- The analysis uses a practical definition of AGI centered on the automation of purely cognitive labor at superior quality and speed.
- Variability in how different experts define AGI remains a significant caveat when interpreting these forecasting trends.