The history of AI is long and varied.
Deploying increasingly capable AI systems
Superhuman automata have been in the human psyche since the advent of thought1. Over time, and with the promulgation of scientific thinking, humanity started looking at the creation of thinking machines as a goal in and of itself2. More recently, we have seen a serious expansion in the capability suite of AI Systems3.
As these systems get deployed, and the associated risks with deployment come to the fore, we must question the fundamentals on which our thinking about AI systems rests upon. As an example, see ideas around assistance games4, where agents are deemed uncertain about specified goals, and elicit human preferences through interaction to then be of value.
Personally, I’ve always favoured the community driven, bottom-up grassroots approach to the crystallisation of ideas. We are at a nascent stage when it comes to economically and culturally valuable AI capabilities making leaps out of labs. The time to steer these in a direction that benefits humanity and the planet is now. Doing so from the bottom up and in the open, driven by an ever expanding crew of enlightened humans, is how I’d like to engage with the nuances around this topic.
Community to steer the AI narrative
The AI Salon meetup group has been created with this mind. The group will promote discussions around the past (incl. in myths), present, and future of AI, with the aim of exploring ideas that feed into engineering the safe deployment of current and future AI systems.
Communities, small or large, are key to shaping culture. Let’s re-examine the roots, and help grow a culture around the safe deployment of AI systems considering all manner of concerns, be they technical, social, legal, and beyond. The open source movement around seriously capable models that has currently gripped the creative imagination of the masses, fueled by Stability AI, as refreshing as it is, makes it even more urgent for the community to develop and drive the techniques and the narrative surrounding safe AI deployment.
I’m looking forward to having you join us.
Generated using Stable Diffusion v1.5 at DreamStudio using the prompt - “Artificial Intelligence safely deployed to augment the human condition and benefit people and the planet”.
Let’s have a conversation, seed collaborations, and help shape the broader narrative around AI to what one feels when one walks along a randomly selected bridge in a randomly selected town on this planet we all call Earth.
Gods and Robots: Myths, Machines, and Ancient Dreams of Technology, Adrienne Mayor ↩
On the Opportunities and Risks of Foundation Models, Center for Research on Foundation Models (CRFM), Stanford University ↩