Uploaded November 2023 | Updated September 2026, 1 week ago
IBM’s Kush Varshney explains what a steam engine controller and AI safety have in common.
Read more from Kush here:
research.ibm.com/blog/AI-governance-explained
Learn more about watsonx.governance
ibm.com/products/watsonx-governance
Subscribe and stay up to date on news and announcements from IBM Research YouTube → http://ibm.biz/subscribe_IBM_Research
For more news, make sure to subscribe to our newsletter, Future Forward:
ibm.com/account/reg/us-en/signup?formid=news-urx-51849
IBM’s Kush Varshney explains what a steam engine controller and AI safety have in common.
Read more from Kush here:
research.ibm.com/blog/AI-governance-explained
Learn more about watsonx.governance
ibm.com/products/watsonx-governance
Subscribe and stay up to date on news and announcements from IBM Research YouTube → http://ibm.biz/subscribe_IBM_Research
For more news, make sure to subscribe to our newsletter, Future Forward:
ibm.com/account/reg/us-en/signup?formid=news-urx-51849



![Why do AI models need to be safe?
IBM Fellow Kush Varshney explains how IBM Research is responding to the evolving safety risks associated with generative AI.
https://research.ibm.com/blog/map-measure-manage-gen-ai
Introduction [00:00]
Why do we need AI safety? [00:09]
Determining what is harmful [1:55]
Granite Guardian to detect risks [3:00]
Alignment vs. steerability [6:17]
How to AI models make decisions? [9:35]
Are certain steering methods more effective? [12:20]
Intrinsic functions for generative AI [16:06]
What is generative computing? [17:05]
Surprises and lessons learned [19:44]
Innovating responsibly [20:58]
Whats next for AI safety research [26:08]
For more news, make sure to subscribe to our newsletter, Future Forward:
https://ibm.biz/BdMdCg
Subscribe to the IBM Research channel → http://ibm.biz/subscribe_IBM_Research
#ai #riskmanagement #generativeai Why do AI models need to be safe?](https://i.ytimg.com/vi/g2A7NUSa7qA/mqdefault.jpg)






