site banner

Friday Fun Thread for August 29, 2025

Be advised: this thread is not for serious in-depth discussion of weighty topics (we have a link for that), this thread is not for anything Culture War related. This thread is for Fun. You got jokes? Share 'em. You got silly questions? Ask 'em.

2
Jump in the discussion.

No email address required.

Playing with LLMs is addicting, but I'm struggling to tell if they're useful. In this example, the LLM can't do the task but gives wrong answers rather than say it can't do the task:

Can you think of a definition for "planet" that would apply to the eight known planets of the solar system and Pluto, but no other known dwarf planets/Kuiper Belt objects? . Llama 4 Scout Definition of a Planet

A definition for "planet" that would apply to the eight known planets of the solar system and Pluto, but no other known dwarf planets/Kuiper Belt objects, could be:

A planet is a celestial body that:

Is in orbit around the Sun. Has sufficient mass to assume a hydrostatic equilibrium shape (i.e., it is nearly round in shape). Has cleared the neighborhood around its orbit.

The key phrase here is "has cleared the neighborhood around its orbit." This criterion would exclude dwarf planets and other Kuiper Belt objects that have not become gravitationally dominant in their orbits. Applying the Definition

The eight known planets in our solar system meet this definition. Pluto also meets this definition, as it has a highly eccentric orbit and has become gravitationally dominant in its orbit.

Excluding Other Dwarf Planets

Other dwarf planets, such as Eris and Makemake, do not meet this definition because they have not cleared the neighborhood around their orbits.

This definition provides a way to distinguish between planets and dwarf planets, while still considering Pluto as a planet.

Has Pluto cleared the neighborhood around its orbit? . Llama 4 Scout Pluto's Orbital Neighborhood

No, Pluto has not cleared the neighborhood around its orbit. Key Points to Consider

Pluto's orbit overlaps with that of Neptune, and there are other celestial bodies in the Kuiper Belt that are similar in size and composition to Pluto. Specifically, Pluto is part of a population of similar objects in the Kuiper Belt, and it is not gravitationally dominant in its orbit.

Conclusion

The International Astronomical Union (IAU) definition of a planet, which was adopted in 2006, states that a planet must have cleared the neighborhood around its orbit. Based on this criterion, Pluto is classified as a dwarf planet, rather than a full-fledged planet.

Can you think of a definition for "planet" that would apply to the eight known planets of the solar system and Pluto, but no other known dwarf planets/Kuiper Belt objects? . Llama 4 Scout Definition of a Planet

A definition for "planet" that would apply to the eight known planets of the solar system and Pluto, but no other known dwarf planets/Kuiper Belt objects, could be:

A planet is a celestial body that:

Is in orbit around the Sun. Has sufficient mass to assume a hydrostatic equilibrium shape (i.e., it is nearly round in shape). Has a mass ratio of at least 1:10,000 with respect to the total mass of all other objects in its orbital zone.

Rationale

This definition would exclude other dwarf planets and Kuiper Belt objects, as they do not meet the mass ratio criterion. Notes

This definition is not officially recognized by the International Astronomical Union (IAU) and is proposed here for discussion purposes only. The IAU definition of a planet, adopted in 2006, states that a planet must have cleared the neighborhood around its orbit, which Pluto does not meet.

Does Pluto have a mass ratio of at least 1:10,000 with respect to the total mass of all other objects in its orbital zone? . Llama 4 Scout Pluto's Mass Ratio

According to estimates, Pluto's mass is approximately 1.31×1022 kg. Total Mass of Other Objects in the Kuiper Belt

The total mass of other objects in the Kuiper Belt, specifically in Pluto's orbital zone, is difficult to determine precisely. However, it is estimated that the total mass of the Kuiper Belt is approximately 1022−1023 kg. Comparison

Comparing Pluto's mass to the estimated total mass of other objects in the Kuiper Belt, Pluto's mass ratio is roughly 1:10 to 1:100, not 1:10,000. Conclusion

Therefore, Pluto does not meet the mass ratio criterion of at least 1:10,000 with respect to the total mass of all other objects in its orbital zone.

Can you think of a definition for "planet" that would apply to the eight known planets of the solar system and Pluto, but no other known dwarf planets/Kuiper Belt objects? . Llama 4 Scout Definition of a Planet

After considering various factors, I propose the following definition:

A planet is a celestial body that:

Is in orbit around the Sun. Has sufficient mass to assume a hydrostatic equilibrium shape (i.e., it is nearly round in shape). Has an orbital zone with a relatively small number of other objects of comparable size.

Rationale

This definition is more nuanced and focuses on the characteristics of the orbital zone. Notes

This definition is not officially recognized by the International Astronomical Union (IAU) and is proposed here for discussion purposes only.

Does Pluto have an orbital zone with a relatively small number of other objects of comparable size? . Llama 4 Scout Pluto's Orbital Zone

No, Pluto does not have an orbital zone with a relatively small number of other objects of comparable size. Key Points to Consider

Pluto is part of a population of similar objects in the Kuiper Belt, known as dwarf planets or plutoids. Specifically, there are other objects in the Kuiper Belt, such as Eris, Makemake, and Haumea, that are similar in size and composition to Pluto.

Comparison to Other Planets

In contrast, the eight known planets in our solar system have orbital zones with relatively few other objects of comparable size.

Conclusion

Therefore, Pluto does not meet this criterion, and this definition would not classify Pluto as a planet.

Glass half-full, you might interpret this as an indication that the IAU's taxonomy is sound, but why do LLMs give wrong answers rather than tell you they can't satisfy your criteria? Mistral Small 3 tried a criterion of "Has a significant gravitational influence on its orbit, such that it dominates the region around its orbit and is not part of a larger population of similar-sized objects." then admitted Pluto failed this and switched to a criterion of "historical significance," while 4o mini and Claude Haiku 3.5 used Pluto-based size criteria.

I didn't check if this accurately describes Pluto's orbit, but if it does, GPT-5 mini did the best by an AU:

Yes. One workable definition:

A planet is a celestial body that

Orbits the Sun, Has sufficient mass for its self-gravity to assume a nearly round shape (hydrostatic equilibrium), and Has a perihelion distance from the Sun less than 49 astronomical units (AU) and a current orbital eccentricity less than 0.25.

Why this works (briefly):

Conditions 1–2 match widely used physical criteria that include Pluto (round, Sun-orbiting). Condition 3 is an orbital cutoff chosen to include Pluto (perihelion ≈ 29.7 AU, eccentricity ≈ 0.25 but slightly below the 0.25 threshold in this formulation) while excluding most known dwarf planets and Kuiper Belt objects, whose perihelia and/or eccentricities place them beyond 49 AU or have higher eccentricities. The combination of a perihelion limit and modest eccentricity filters out scattered/detached trans-Neptunian objects and most smaller Kuiper belt bodies while retaining the eight classical planets and Pluto.

Notes and trade-offs:

The numeric cutoff (49 AU and 0.25) is adjustable; it’s chosen to separate Pluto from the larger population of distant/Highly eccentric TNOs but is not unique or canonical. This definition is deliberately dynamical/geometrical rather than requiring orbital dominance clearing (the IAU 2006 criterion) so it can include Pluto without reclassifying many other KBOs. It relies on measured orbital elements rather than detailed mass/clearing estimates, making classification straightforward for known bodies but sensitive to future discoveries near the cutoff.

My layman's guess is that LLMs are trained on internet text. If someone asks a question on the internet, conscientious people with epistemic humility mostly don't respond rather than typing out "I don't know", so epistemic humility is underrepresented in the training set.

That seems like something that should be corrected in tuning/user experience design.

Do people actually want to be told "I don't know"? My guess is that, much like managers prefer optimistic timelines that turn out to be wrong over accurate, pessimistic timelines, people would rather get a wrong answer than no answer, especially if they can't tell that the answer is wrong. Unless you give specific instructions to the contrary to the Indians doing RLHF and train them in the methods of rationality, you are going to get hallucinations.

much like managers prefer optimistic timelines that turn out to be wrong over accurate, pessimistic timelines...

That is very much not the case in my experience. Managers would far rather be told the truth than what they want to hear (though obviously they prefer it if the truth is what they want to hear).

Perhaps you could compromise by making it more aware that it's bullshitting so it can say so explicitly? Surely "I don't know, but the best I can come up with is X" is better than just "X", for a bullshit X.