OpenAI launched its first ChatGPT Futures class, granting 26 young builders $10,000 each and access to frontier models. One recipient, 18‑year‑old high school student Matteo Paz, applied machine learning to nearly 200 TB of NEOWISE infrared data and flagged about 1.9 million variable objects, roughly 1.5 million of which appear to be previously unknown candidates.
What the researcher did
Matteo Paz processed a very large NEOWISE infrared dataset using machine learning techniques. The dataset he worked with totaled nearly 200 terabytes, and his analysis identified approximately 1.9 million objects exhibiting variability in brightness. Of those, around 1.5 million are described as candidate objects that were not previously catalogued in the shared data.
Why this matters
The notable point is that a high‑school student — equipped with mentors, compute, and access to modern AI models — tackled a complex, large‑scale scientific problem. This example suggests the pathway into research is shifting away from the traditional stepwise academic ladder (PhD, postdoc, institutional lab access and permissions). The barrier has not disappeared but shifted: success increasingly depends on problem framing, taste, and execution rather than formal credentials alone.
Implications for the research ecosystem
- Broader availability of resources (mentorship, compute, models) means talent may emerge outside formal credentialing systems.
- Universities and labs should monitor who is working on relevant problems, because significant contributions may now come from nontraditional entry points.
- The change increases competitive pressure: individuals with sufficient dedication and access can become immediate contenders in frontier research.
Conclusion
The first ChatGPT Futures class and Matteo Paz’s work illustrate how access to AI and computing resources can reshape early stages of scientific investigation. Research institutions may need to reconsider how they identify and engage emerging talent, since influential contributors may arrive without the conventional academic trajectory.
