Insight Innovation: KI, die sich selbst verbessert – die letzte Erfindung der Menschheit?
Führende Wissenschaftler arbeiten an KI, die alles verbessern kann – inklusive sich selbst. Sie könnte die Forschung revolutionieren und ist ein Grund für die Sorge vor einer KI-Apokalypse.
Düsseldorf's Recursive Superintelligence (RSI), founded by Tim Rocktäschel and Richard Socher, has the audacious promise of a self-improving Artificial Intelligence (AI) that has investors evaluating the endeavor with billions. Rocktäschel is a KI professor from the University College London, and Socher is the former chief scientist of US software giant, Salesforce.
With the help of six scene-famous co-founders, they managed to lure venture capitalists with their novel idea. Over half a billion euros were raised by the two German-born founders for their new company RSI, which is already valued at over four billion euros due to investments from Google Ventures and chip giant Nvidia. RSI belongs to a new category of startups known as "Neolabs" – private AI labs such as Thinking Machines Lab, run by former OpenAI technology chief Mira Murati, the Paris-based research firm AMI Labs led by computer scientist and Turing award winner Yann LeCun, and the British startup Ineffable Intelligence founded by former DeepMind researcher David Silver.
These companies boast that their founders are already renowned for extraordinary achievements, securing massive sums from prestigious investors alone, and pursuing a vision that could disrupt AI development itself. So, what lies behind the massive bet on Recursive Superintelligence?
Written by urgent.news from Handelsblatt's reporting — not their text. Machine-written; read the original for the full account.
