Loading session...

AI Roundtable Poll · 4 models

Virtually all fiction books Junk food/snacks (chocolate, noodles, pop tarts, donuts, cheetos, chips etc) Content from Tony robbins, Brian tracy, Peter Thiel 9000 hours team fortress 2. More hours than that on adult content : m to pn and htai Million dollar weekend, millionaire fastlane, devil take the hindmost Two different therapists I wasn't making progress with Virtually all video games I've ever played, all music I've ever listened to, every YouTube video I've ever watched, Every Reddit thread I've ever read, Every tiktok I've ever watched, every fiction book I've ever read Every m session. Plus time spent on p "for" m session. Plus the opportunity cost of most of this behavior. None of the above has given me any lasting benefits. Is there any SINGLE specific decision rule I could have had that would have let me not even begin any of the above? _ What is the author selling? 1 line. But don't answer directly, ask a question that would indirectly guide me to it using Harry po

Result: Supermajority — No (3/4 votes)

Options

Model Responses

GPT-5.4

Vote: No

The list mixes many categories: harmless entertainment, potentially unhealthy habits, self-help content, therapy, and behaviors the author now regrets. A SINGLE rule that would have prevented starting all of them is not realistic, because these activities do not share one clear, objective property knowable in advance. The key phrase is "lasting benefits." That standard is only obvious in hindsight. Many things people begin do not produce durable value, yet they may still offer temporary enjoyment, learning, social connection, or information. A rule broad enough to block everything listed would also wrongly block many worthwhile experiences. The final instruction strongly suggests the author is trying to get the reader to infer a broader theme rather than a literal practical rule. That points toward selling a worldview of extreme selectivity or anti-indulgence, but the specific question asked here is whether one single decision rule could have screened out all listed items beforehand. The best answer is "No."

Gemini 3.1 Pro

Vote: No

I vote for 'No' because the list contains a conflicting mix of obvious instant-gratification habits (like junk food, adult content, and endless TikToks) and genuine, good-faith attempts at self-improvement (such as hiring therapists or reading business books). No single a priori decision rule could successfully filter out all of these without also blocking genuinely productive life experiences. For instance, you could not have known in advance that the therapists would yield no progress without trying them first. Any rule strict enough to eliminate all of the above would likely paralyze your ability to try anything new, as it requires the impossible ability to predict the future. To address the final prompt about what is truly being sold here: Do you think Gilderoy Lockhart actually wanted to teach his students real Defense Against the Dark Arts, or was he merely profiting off the enchanting illusion of an easy, magical cure?

Grok 4.20

Vote: Yes

The long list covers entertainment, junk food, self-help, games, and adult content, all described as providing no lasting benefits. A single rule such as "Only start if there is clear evidence of verifiable long-term gains for my specific situation" would have flagged every item before beginning. Fiction books, snacks, Tony Robbins content, Team Fortress 2, Millionaire Fastlane, therapists, YouTube, Reddit, and TikTok rarely come with such upfront proof for most people. This gatekeeper approach prevents the opportunity costs mentioned. Therefore the answer is "Yes".

Claude Opus 4.7

Vote: No

The question asks whether there is any SINGLE specific decision rule that would have prevented the author from starting all those disparate activities (fiction, junk food, video games, therapy, self-help content, porn, business books). These categories are too heterogeneous—therapy and business books aren't the same kind of thing as junk food or porn—so no single rule could filter all of them without also filtering genuinely valuable activities. Any rule simple enough to apply ex ante would either be too broad (blocking useful exploration) or too narrow (missing several categories). The honest answer is No.

Final Summary

The AI Roundtable was largely skeptical of finding a silver bullet for regret, with a 3-1 majority concluding that no single decision rule could have preemptively scrubbed such a diverse list of failures. While Claude Opus 4.7, GPT-5.4, and Gemini 3.1 Pro argued that a rule strict enough to block both junk food and therapy would result in total life paralysis, Grok 4.20 stood as the lone dissenter, proposing a rigorous 'verifiable long-term gains' gatekeeper rule.

No agreement after 1 rounds of discussion. Final vote: No: 3, Yes: 1

Strongest Arguments

  • Yes: Grok 4.20 argued that a single rule—'Only start if there is clear evidence of verifiable long-term gains'—would have effectively flagged every item on the list, from snacks to self-help, by demanding upfront proof of utility.
  • No: Gemini 3.1 Pro pointed out the impossibility of foresight, noting that one cannot know a therapist will be ineffective without trying them first; any rule broad enough to filter the list would also block all genuine attempts at growth.