SingularityNET is betting on a network of highly effective supercomputers to get us to Synthetic Common Intelligence (AGI), with the primary one set to whir into motion this September.
Whereas as we speak’s AI excels in particular areas – suppose GPT-4 composing poetry or DeepMind’s AlphaFold predicting protein buildings – it’s nonetheless miles away from real human-like intelligence.
“Whereas the novel neural-symbolic AI approaches developed by the SingularityNET AI group lower the necessity for knowledge, processing and power considerably relative to commonplace deep neural nets, we nonetheless want important supercomputing services,” SingularityNET CEO Ben Goertzel defined to LiveScience in a latest written assertion.
Enter SingularityNET’s formidable plan: a “multi-level cognitive computing network” designed to host and practice the extremely complicated AI architectures required for AGI. Think about deep neural networks that mimic the human mind, huge language fashions (LLMs) educated on colossal datasets, and techniques that seamlessly weave collectively human behaviours like speech and motion with multimedia outputs.
However this stage of sophistication doesn’t come low-cost. The primary supercomputer, slated for completion by early 2025, shall be a Frankensteinian beast of cutting-edge {hardware}: Nvidia GPUs, AMD processors, Tenstorrent server racks – you identify it, it’s in there.
This, Goertzel believes, is greater than only a technological leap, it’s a philosophical one: “Earlier than our eyes, a paradigmatic shift is going down in direction of steady studying, seamless generalisation, and reflexive AI self-modification.”
To handle this distributed network and its treasured knowledge, SingularityNET has developed OpenCog Hyperon, an open-source software program framework particularly designed for AI techniques. Consider it because the conductor making an attempt to make sense of a symphony performed throughout a number of live performance halls.
However SingularityNET isn’t retaining all this brainpower to itself. Harking back to arcade tokens, customers will buy entry to the supercomputer network with the AGIX token on blockchains like Ethereum and Cardano and contribute knowledge to the collective pool—fuelling additional AGI improvement.
With consultants like DeepMind’s Shane Legg predicting human-level AI by 2028, the race is on. Solely time will inform if this world network of silicon brains will start the subsequent nice leap in synthetic intelligence.
(Photograph by Anshita Nair)
See additionally: The merging of AI and blockchain was inevitable – but what will it mean?
Need to study extra about AI and massive knowledge from business leaders? Try AI & Big Data Expo going down in Amsterdam, California, and London. The excellent occasion is co-located with different main occasions together with Intelligent Automation Conference, BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.
Discover different upcoming enterprise expertise occasions and webinars powered by TechForge here.