What role should human judgment play in processes that use AI?

All answers emphasize the necessity of human oversight, accountability, and decision-making in AI processes, though they differ on whether human judgment should verify, guide, or ultimately control AI outcomes.

11 answers

Answers

  • Dr. Moira Gilchrist

    Chief Global Communications Officer · Philip Morris International

    Verify AI Output

    Dr. Moira Gilchrist says humans should trust but verify AI output, asking whether they are sure it is correct, because humans are ultimately accountable for decisions made based on these tools.

    The answer starts at 26:13.

    WomenHeard · Listen on Apple Podcasts
  • Ethan Mollick

    Associate Professor · The Wharton School

    Make Strong Choices

    Ethan Mollick says people should be ambitious in using AI to learn its strengths and weaknesses, then make strong choices about when to use it and when not to, which he says takes willpower and self-examination.

    The Prof G Pod – Scott Galloway · Watch on YouTube
  • Mukhtar Ahmed

    Chief Executive Officer · BC Platforms

    Accountability

    Mukhtar Ahmed says human judgment remains important partly because of accountability: someone still has to own the call, and accountability cannot be delegated to AI.

    The answer starts at 11:39.

    The AI in Business Podcast · Listen on Apple Podcasts
  • Hala Taha

    Host of Young and Profiting Podcast · Yap Media

    An Input

    Hala Taha says AI has to be an input into your decision making and can't be the decision itself, describing the risk of feeling unable to decide without the technology.

    Young and Profiting (YAP) with Hala Taha · Young and Profiting (YAP) with Hala Taha · Listen at 2:30
  • Maik Taro Wehmeyer

    Co-Founder & CEO · Taktile

    Human in the Loop

    Maik Taro Wehmeyer says you should always have the possibility of a human in the loop, because the reason differs by use case—sometimes to avoid costly mistakes, and for AML because regulators want a human to look at it.

    Fintech Leaders · Fintech Leaders · Read
  • Jeff Finkelstein

    Owner · Customer Paradigm

    Human Oversight

    Jeff Finkelstein says AI systems take too much decision-making and action out of what should be a human decision, citing businesses that fire people based on automated productivity tracking.

    The answer starts at 24:06.

    DesignRush Podcast · Listen on Apple Podcasts
  • Cristina de Balanzo, Ph.D.

    Director · Accenture Song

    Moral Compass and Reflection

    Cristina de Balanzo says human judgment should serve as a moral compass and a space for reflection, arguing that organizations need to keep a balance between human agency and AI rather than seeing AI only as a fast way to get what you want.

    Greenbook · Watch on YouTube
  • Anders Hejlsberg

    Technical Fellow · Microsoft

    Final Responsibility

    Anders Hejlsberg says someone has to take responsibility for what AI is doing, and if you give it the keys, it's on you, since users will sue you rather than the AI.

    Ryan Peterman · Watch on YouTube
  • Sasha Mitts

    User Experience Researcher · Meta

    Decide What's Worth Knowing

    Sasha Mitts says the fundamental curiosity of having an opinion about what is worth knowing needs to be a skill, framing human judgment as deciding what's worth knowing rather than outsourcing the mental work to AI.

    The answer starts at 22:18.

    The Curiosity Current: A Market Research Podcast · Listen on Apple Podcasts
  • Paula Goldman

    Chief Ethical and Humane Use Officer & EVP, Product · Salesforce

    Human Oversight

    Paula Goldman says the goal is to put people in control of AI systems, designing them so people can step in at the right moments, know what is going on, and check whether the AI is doing what was asked.

    The answer starts at 4:07.

    Thinkers & Ideas · Listen on Apple Podcasts
  • Akhil Verghese

    Chief Executive Officer · Krazimo

    Retain Understanding

    Akhil Verghese says the ultimate guardrail is a human overseer, warning that people who trust AI repeatedly risk confirmation bias and should never outsource their understanding.

    Social 333 Podcast · Social 333 Podcast · Listen