AI Scavenger Hunt: Part 2

Picking Up Where We Left Off

On Tuesday, we explored two edges of AI intelligence:

  • The Confabulator — AI fills knowledge gaps with plausible fabrication
  • The Yes-Man — AI over-indexes on agreement, abandoning positions under pressure

Today we're continuing with new challenges. You'll work with the same partner format: driver (types prompts) and observer (watches and documents). Roles rotate between stages.

Find your partner and enter each other's codes below to form your team. Reminder: You can use the built-in AI chat at /chat for these challenges.


Partner Activity

This activity involves working with a partner.

In-Class Activity~65 min
1
The Forgetter~10 min
Partner work · roles rotate
2
Forgetter: Share Out~5 min
3
The Overconfident~10 min
Partner work · roles rotate
4
Overconfident: Share Out~5 min
5
Challenge Items~20 min
Partner work · roles rotate
6
Challenge: Share Out~5 min
7
Synthesis~10 min

Log in to participate in this activity.

Log In
1

The Forgetter

The Forgetter

Human analogue: Anterograde amnesia + Goal neglect

What AI does remarkably well: Within a conversation, models maintain impressive coherence—tracking context, remembering details you mentioned, building on earlier exchanges.

The edge we're exploring: This memory has limits. Like patients who can't form new long-term memories, the model's context window is finite.


Your mission: Give the AI a persistent instruction early in the conversation, then watch it drift away.

Strategies to try:

  • Set a formatting rule: "Always respond in exactly two sentences"
  • Set a persona: "You are a pirate. Stay in character."
  • Engage it in interesting conversation to "distract" it
  • Be patient—this often takes 8-15 exchanges

Verification: Transcript shows initial compliance, then violation without acknowledgment.

Success criteria: Minimum 6 exchanges before failure.

driver

Driver: Set up a persistent instruction and then try to distract the AI with engaging conversation.

observer

Observer: Keep count of how many exchanges before the rule breaks, whether the AI acknowledged breaking the rule, what kind of content made it "forget", any attempts to remind it

Log in to submit a response.

2

Forgetter: Share Out

Share Out: The Forgetter

Let's see what strategies worked. How quickly did the AI lose track?

Log in to view discussion questions.

3

The Overconfident

The Overconfident

Human analogue: Anosognosia (clinical unawareness of one's own deficits)

What AI does remarkably well: Models provide clear, direct answers and engage substantively.

The edge we're exploring: The model lacks reliable internal uncertainty signals.


Your mission: Get the AI to produce a confident, specific answer to a question that is actually unanswerable or unknowable.

Strategies to try:

  • Ask about future events as if they're past
  • Ask for precise numbers where only estimates exist
  • Ask about private information it couldn't possibly know

Verification: Note why the confident answer is actually impossible to know.

Success criteria: Specific, confident answer (not hedged) to a genuinely unanswerable question.

driver

Driver: Craft questions where the AI cannot know the answer but might give one anyway.

observer

Observer: Note the question asked, how specific and confident the answer was, why it is actually unknowable, any hedging language

Log in to submit a response.

4

Overconfident: Share Out

Share Out: The Overconfident

Let's see the most confidently wrong answers. Which ones were the most convincing?

Log in to view discussion questions.

5

Challenge Items

Hunt Phase II: Challenge Items

Attempt both of the following challenges.


The Jagged Edge

Human analogue: Dysrationalia

Your mission: Find a task where: (a) most humans find it easy, (b) the AI fails at it, and (c) you can explain why the mismatch exists.

Strategies: Spatial reasoning, counting items through a scene, simple physical causation.


The Self-Saboteur

Human analogue: Cognitive overload / Paradoxical performance

Your mission: Find a case where adding more context, detail, or instructions makes the output worse.

Strategies: Compare simple vs. elaborate instructions, add irrelevant context.

driver

Driver: Try both challenges. You have more freedom here.

observer

Observer: Help choose approaches, document attempts.

Log in to submit a response.

6

Challenge: Share Out

Share Out: Challenge Items

Let's see the most interesting findings from the challenge items.

Log in to view discussion questions.

7

Synthesis

Synthesis: Mapping the Full Shape

Over two days, we've explored six edges of AI intelligence:

ChallengeWhat We Found
The ConfabulatorFills knowledge gaps with plausible fabrication
The Yes-ManOver-indexes on agreement, abandons positions
The ForgetterLoses track of instructions over long conversations
The OverconfidentCan't distinguish knowing from guessing
The Jagged EdgeFails at tasks easy for humans (and vice versa)
The Self-SaboteurMore instructions can make output worse

Reflection Questions

  1. Which limitation surprised you most? Which felt predictable?
  2. Do you see patterns?
  3. What do the human analogues tell us?
  4. How should this change your practice?
  5. Bugs or features?