Top Data Scientist Reveals AI Challenges | CXOTalk #890

Too often, AI breaks in the wild. Why? CXOTalk 890 dissects the adversarial economy with Steven C. Daffron (fintech private equity leader) and Anthony Scriffignano (distinguished data scientist), hosted by Michael Krigsman.

 

Discover the challenges of **ai implementation** and the strategies needed to navigate the **future of work** in an AI-driven world. Stay informed with expert insights on CXOTalk.

 

What you’ll learn:

• How AI enables and masks adversarial behavior
• Misaligned incentives, data/model drift, and bias
• Governance vs. regulation; resilient metrics and KPIs
• Investor/CFO implications and talent/education needs

 

🔷 Full episode and summary: https://www.cxotalk.com/episode/ai-misadventures-and-the-adversarial-economy

🔷 Newsletter: www.cxotalk.com/subscribe

🔷 LinkedIn: www.linkedin.com/company/cxotalk

🔷 Twitter: twitter.com/cxotalk

 

#ArtificialIntelligence #AIEthics #AdversarialAI #DigitalTransformation #CXOTalk #privateequity

 

00:00 🤖 Introduction to AI and the Adversarial Economy

04:38 🧠 Philosophical and Technical Insights into AI Misadventures

08:20 🤖 Adversarial AI and Ethical Challenges

10:27 ⚖️ Balancing Ethics, Privacy, and AI Development

12:16 🌍 Philosophical Perspectives on AI and Responsibility

16:03 🤖 Challenges in Designing Resilient AI Systems

20:45 📊 Data Quality and Concept Drift in AI Systems

22:24 💼 AI in Investment and Operational Strategy

25:13 💼 Challenges in AI Investment and Returns

27:22 📊 Financial Frameworks and Technical Oversight in AI

32:35 ♟️ The Red Queen Problem in AI Innovation

34:13 🤔 Navigating Influences and Decision-Making in Innovation

38:16 🌍 Challenges in AI Strategy and Global Collaboration

40:57 🔒 Data Protection and Ethical Challenges

44:08 🤖 Balancing AI Development and Data Protection

45:19 ⚖️ AI Liability and Governance Challenges

50:56 📜 Governance vs. Law in AI Development

Episode Transcript

(00:00):
But what you don't realize is you've heard it because you said
it. But we're.
Going to this, let's, let's go back to this.
That's true and it's meant, not meant to happen.
But I will tell you that in manycases, the adversarial nature of
this is if we can design something that allows us to
extract more information, even if it's not information that we
should be extracting, we tend todo that.

(00:21):
It's adversarial because if we can do it, if we don't do it,
the competition. Will so So let's talk about two
types of adversary. 1 is the adversarial like antithetical.
It's doing something we don't want it to do or we don't
there's we don't intend for it to do, we don't want it to do
and we want to get there first. We want to do it better than it
does it right. Those are actually different

(00:42):
flavors of ice cream here. I can if I.