Can we align AI before its too late?

​AI grows more capable every day. Can we make it reliably do what we intend before it causes serious harm? Join us for an educational session on the AI alignment challenge. Attendees will hear an introduction to the problem of AI alignment and the risks that could arise if alignment isn’t achieved, including:

• ​Supply-chain emergencies

• ​Biosecurity threats

• ​Loss of human control

​This will be followed by a panel discussion, a Q&A and small-group discussions at attendees’ tables on how AI alignment can help bring about a positive future for humanity and AI. We’ll explore what needs to be done to achieve alignment, and how likely it is to be achieved in time to avoid significant risks to human flourishing.

​Speakers

• ​Daniel Tan, AI Safety Researcher, UCL

• ​Joseph Miller, Director, PauseAI UK

• ​Siliconversations, educational influencer

​Timeline

• ​6:00pm: Meet and greet

• ​6:30pm: Introduction to the AI alignment challenge

• ​7:00pm: Panel discussion and Q & A

• ​8:00pm: Interactive Games

• ​8:20pm: Small-group discussions

​Co-hosted by PauseAI UK and the Bentham Brooks Institute, a UCL student-led think tank.