Is AI a Threat? Professor Shannon Vallor

Why It Matters

Recognizing effective AI governance tools counters fatalistic narratives, enabling policymakers and firms to manage risks and harness AI’s commercial potential responsibly.

Key Takeaways

• Techno‑fatalism ignores historical ability to regulate disruptive tech.
• Society has successfully rejected unsafe technologies despite commercial pressure.
• Existing AI governance tools include model cards, audits, and procurement standards.
• Assurance cases can certify AI safety and fitness for purpose.
• Ethical AI frameworks are already operational, countering fatalistic narratives.

Summary

Professor Shannon Vallor challenges the growing narrative that artificial intelligence is an unstoppable force beyond human control. She argues that techno‑fatalism contradicts centuries‑long evidence of societies regulating, restricting, or even abandoning technologies deemed hazardous, regardless of their commercial allure. Vallor highlights concrete mechanisms already in place to govern AI ethically. Model cards, algorithmic audits, and procurement standards provide transparency, while assurance cases offer formal certification of safety and fitness for purpose. These tools demonstrate that the industry possesses a robust toolbox for responsible AI deployment. She cites specific examples—model cards that disclose performance metrics, audits that detect bias, and procurement clauses that enforce ethical criteria—as proof that governance is not theoretical but actively practiced. Assurance cases, borrowed from safety‑critical engineering, further illustrate how AI systems can be rigorously vetted before release. The implication for businesses and regulators is clear: dismissing AI as an uncontrollable threat ignores existing governance frameworks and undermines proactive risk management. Embracing these tools can steer AI development toward societal benefit rather than fatalistic resignation.

Original Description

Professor Shannon Vallor describes technofatalism, the idea that technology will always be a threat.

Hear more from Professor Vallor at our upcoming panel event: https://royalinstitutephilosophy.org/event/is-ai-a-threat/