top of page
Writer's pictureAIIA

David Shapiro: Can We Build AI Without Losing Control Of It?




David Shapiro kicks off his talk, "Can We Build AI Without Losing Control Over It?" with a laid-back yet insightful dive into the world of artificial intelligence (AI). Right from the get-go, Shapiro tackles the big question hanging over everyone's heads: How do we keep the reins on machines that might end up being smarter than us?


This concern, known in the tech world as the control or alignment problem, is chewed over from every angle, mixing tech talk with some deep philosophical musings.


On the tech side of things, Shapiro gets into what's called inner alignment, breaking down the idea into something like making sure AI systems are actually aiming for what we want them to. He uses the example of how schools grade students to show how easy it is to end up with results we didn't intend, thanks to the way we measure success.

Then, the conversation takes a turn to outer alignment, pondering if the goals we're setting for AI are actually good for us and the planet. It's here that Shapiro asks the tough questions: What values do we program into AI? And how do we make sure these goals are in line with what's best for everyone?


Captain Shapiro doesn't shy away from the fast pace at which AI is evolving, sparking a chat on the possibility of AI outsmarting human brains. This opens up a can of worms on the risks involved with AI, like the unexpected outcomes and how we're supposed to handle these challenges as technology races ahead.


The talk isn't all doom and gloom, though. Shapiro emphasizes the silver lining—research, education, and open conversations as tools to mitigate AI risks. He's all for a team effort, calling on everyone from techies to policymakers to pitch in and help steer AI development in a direction that's beneficial for humanity.


Packing it all up:

David's chat serves as a beginner-friendly guide on the knotty problem of keeping AI in check as it grows more complex.


He highlights the critical need to align AI's goals with human values and underlines the importance of continuous research and dialogue to navigate the tricky waters of AI technology and its impact on our world. Please watch the video.

And be happy. :D



31 views0 comments

Recent Posts

See All

Comments


bottom of page