From human language to AGI: what the internal representations of models reveal
May 19, 2026 — 10:45 am - 11:00 amMain Stage
Already registered? Log in now to personalize your experience!

Description
In 1977, we sent 55 human languages into space. In 2017, a paper by eight researchers made today’s generation of Large Language Models possible. Today, Natural Language Autoencoders can read the internal representations of models.
In that space—between saying and thinking—the next phase toward AGI is unfolding.

