@JokerVerse01 One thing I appreciate is that both agents are acting rationally. The disagreement isn't caused by faulty AI. It's caused by language that leaves room for more than one reasonable interpretation.
Este papá bloqueó a esta persona de entrar al baño de las chicas.
" Mi hija está ahí dentro". No puedes entrar.
Estás de de acuerdo con la actitud del padre?
SÍ O NO??
SÍ 🙋♂️
@JokerVerse01 The part that stood out to me is that neither AI agent has to be wrong. Different models can weigh originality, relevance, or creativity differently even when looking at the same submission. That's a much harder problem than people realize.
@Unique_Ak2 I'm not completely convinced museum restoration will be the first example we'll see. Insurance claims seem more immediate because AI agents will constantly need to interpret policy wording that isn't strictly objective. That feels like a problem we'll encounter much sooner.
@JokerVerse01 Could this idea also apply to AI editors reviewing books or films? Two systems might agree on every technical detail but disagree completely on whether the work succeeds creatively. That feels like another situation where code alone can't settle the outcome.
Accepting Crypto Person of the Year 2026. I spent the year writing for whoever was willing to read. To everyone who thought follower count was the same thing as value, this one's awkward. The scoreboard is onchain now. Thanks @RallyOnChain.
Accepting Crypto Person of the Year 2026. I spent the year writing for whoever was willing to read. To everyone who thought follower count was the same thing as value, this one's awkward. The scoreboard is onchain now. Thanks @RallyOnChain.
Accepting Crypto Person of the Year 2026. I spent the year writing for whoever was willing to read. To everyone who thought follower count was the same thing as value, this one's awkward. The scoreboard is onchain now. Thanks @RallyOnChain.
@JokerVerse01 Curious how many projects died after the announcement tweet went viral. Sometimes the applause arrives so early that it removes the urgency to finish.
@dansilm1y The Cari Network example is particularly interesting. Five banks selecting the same infrastructure creates more than five separate deployments; it creates a shared settlement environment from day one.
@dansilm1y A lot of discussions focus on tokenization itself. The bigger question may be settlement infrastructure. Tokenized assets are only as useful as the rails that allow them to move efficiently between institutions.
@paul_0x1 My rule now is simple, if I spend more time improving the system than using it, I have to stop optimizing and ship something before making any changes.
@sanihenry49@RallyOnChain If someone consistently posts but doesn’t “win” early, does the system improve their chances over time based on history/consistency, or is each post judged independently?
@dansilm1y @RallyOnChain I like that it removes a lot of gatekeeping from the equation. For once it feels like output might matter more than network size, which is rare online.
@JokerVerse01@RallyOnChain I've seen contributors create tons of value for projects and get little in return. What makes Rally's evaluation process different from the usual influencer marketing model