AI for Faculty and Admin

Colin Madland

2025-06-12

Once Upon a Time…

cmad.land - slides

Why is genAI compelling?

anthropomorphosis

  • we naturally pretend that there is agency behind seemingly sensible outputs

What is intelligence?

What is genAI doing?

REGRESSION

PROBABILITY

genAI and the truth

genAI as a mirror

  • the only basis for an output is historical data with all its embedded biases,
  • …but amplified and distorted

What do mirrors not show?

Picture of a white man’s legs standing in a small river.

HALLUCINATIONS

  • genAI is not capable of generating ‘hallucinations’, which are errors in perception
  • genAI contains no mind with which to perceive anything

LIES

  • likewise, genAI is not capable of lying, which requires a knowing departure from the truth
  • genAI has no mind with which to apprehend the truth of anything nor to exercise agency

“bullshenanigans”

  • bullshenanigans do not require any grounding in reality, it is simply extruded to fill a gap, or ‘flood the zone’ (Frankfurt, 2005)

Sources of BS (Ji et al., 2023)

  • bad data
  • bad inferences from training
  • bad encoding
  • previous BS

Mitigating BS

  • good data
  • good training
  • good encoding
  • there’s nothing you can do, BS is inevitable

Dealing with BS

  • human fact-checking
  • having two chatbots respond to each other’s output until they align (!?)

\(p(\text{doom})\) vs \(p(\text{hope})\)

Recommendations

  • genAI is not inevitable.
  • Resistance and Refusal are possible.
  • small, focussed
  • pay attention to inputs and outputs

Input

  • text
  • music
  • image

Output

  • translated text
  • transposed music, chord charts, separated tracks
  • arrest warrant

Resources

colin.madland@twu.ca | https://cmad.land

Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜. Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610–623. https://doi.org/10.1145/3442188.3445922
Bender, E., & Hanna, A. (2025). AI CON: The hype, the myth and the power grab of the century. The Bodley Head.
Frankfurt, H. G. (2005). On bullshit. Princeton University Press.
Ji, Z., Lee, N., Frieske, R., Yu, T., Su, D., Xu, Y., Ishii, E., Bang, Y. J., Madotto, A., & Fung, P. (2023). Survey of Hallucination in Natural Language Generation. ACM Computing Surveys, 55(12), 1–38. https://doi.org/10.1145/3571730
Vallor, S. (2024). The AI mirror: How to reclaim our humanity in an age of machine thinking. Oxford University press.