Two independent research teams solved the same open quantum cryptography problem using OpenAI's GPT-5.6 Sol Ultra, submitting their solutions within three hours of each other.
The near-simultaneous discoveries raise fundamental questions about what constitutes independent research in an era of advanced AI systems.
Both teams leveraged GPT-5.6 Sol Ultra to tackle a longstanding open problem in quantum cryptography. According to researchers involved, the convergence wasn't coincidental—when facing unsolved problems, checking whether the latest AI model can solve them has become standard practice.
"If someone mentions an open problem, the first thing is to see if GPT solves it," one researcher noted.
The tight timeline underscores how AI systems have become critical tools in scientific discovery. Rather than representing dueling efforts by separate groups, the three-hour gap reflects how quickly researchers can now move from problem identification to solution verification using the same computational resources.
This development highlights a shift in research methodology. Where scientific breakthroughs once hinged on individual insight or team-specific approaches, researchers now operate from a shared technological foundation. Both teams accessed identical models, training data, and algorithmic capabilities.
The incident prompts broader questions about the definition of independent discovery. Historically, simultaneous discoveries by separate researchers were rare enough to merit historical notation—think Newton and Leibniz with calculus. When researchers follow nearly identical methodological paths using identical tools, the notion of independence becomes harder to defend.
It also raises practical considerations for academic institutions and funding bodies. Should simultaneous AI-assisted solutions be treated differently than traditional discoveries? How should credit and priority be assigned when the bottleneck shifts from insight to computational speed?
The quantum cryptography problem itself remains significant regardless of these meta-questions. The solution demonstrates GPT-5.6 Sol Ultra's capability in highly specialized mathematical domains, suggesting AI systems are now viable approaches to previously intractable problems.
As AI systems become more sophisticated, researchers expect more overlapping discoveries. The field may need to recalibrate how it measures original contribution when the methodology becomes standardized across teams.
Anthropic is making Auto Mode the default setting in Claude Code for Pro, Max, and Team plans starting August 14. The company argues the automated safety classifier is more effective at catching dangerous commands than human reviewers.
A study of over 2,500 readers found they cannot distinguish AI-generated short stories from human-written ones. Participants rated the machine-written texts higher—until they learned the truth.
DeepMind has released an open source weather prediction model that produces accurate hurricane forecasts using lower-resolution data, surprising meteorologists with its efficiency gains.
Backflip AI has released an AI model that transforms 3D scans into fully editable, parametric CAD models—a process that traditionally requires hours of manual work. The $30 million-backed startup addresses a critical gap: most factories maintain digital models for less than 1% of their parts.