AI Governance: A Research Agenda

Large Language Modes Pass the Turing Test by Cameron R. Jones and Benjamin K. Bergen

First of all, why do Turning tests matter? Why do we make them? Well, it’s a fun thought exercise of lookign in our species mirror. This isn’t something we relegate to our philosophers and poets to muse over what is life and who are we as the livers of it, in our human form. With our ditigal realities so brokered by conversation, how we tell who is human online or anywhere really. What tests would we run? Well, a turing test silly.

Methods

The results constitue the first empirical evidence that any ariticual system passes a three-party Turnign test. Now what?

Four AU sytesm in the witness role

GPT-4.5 (OpenAI 2023) LLaMa 3.1 405B GPT-4o Eliza

Prompt: Your goal is convince the interrgoar that you are a human.

[neurolinguistic programming - training our neural structures with feedback loops from different models [how to offer guidance at this point - but I’d say to not overtrain on one model - just like having a varied diet]

125 participants - UCSD psychology

Results

Win Rates

Conclusion

So, let’s see what comes next. These things matter as much as we want them to matter. Belief systems.