Hardik Upadhyay

Engineering & Technology Consultant

[Hardik Upadhyay @Clear4C]

@YTJustGetATesla @roadsterbobby I think you should try to overlap multiple trendlines (technological progress of ml & nn, acceptance rate, reoccurring revenue, etc.) of just go look at people who were defending blackberry's keyboard and Nokia's physical strength (after vonerable iPhone was released)

[Hardik Upadhyay @Clear4C]

@PhysInHistory Boring patents are great sleeping medicine

[Hardik Upadhyay @Clear4C]

https://t.co/AUjx5WpTAX

[Hardik Upadhyay @Clear4C]

@theepicmap Can you make this 50/50 globally too? GDP

[Hardik Upadhyay @Clear4C]

@elonmusk @karpathy Matter flow matters to photons 🤔

[Hardik Upadhyay @Clear4C]

@karpathy Carbon Energy vs Silicone Electricity vs ? (Thermonuclear 💥 Explosion, Black hole🕳️, Darkeven Matter, quantum craziness... or what's your best guesses?)

[Hardik Upadhyay @Clear4C]

@karpathy On an exponential curve, if you were to extrapolate the biological evolution of the human brain with current electrical/software based LLM 'intelligence'... What would the new system features in the future look like?

[Andrej Karpathy @karpathy]

Something I think people continue to have poor intuition for: The space of intelligences is large and animal intelligence (the only kind we've ever known) is only a single point, arising from a very specific kind of optimization that is fundamentally distinct from that of our technology.

Animal intelligence optimization pressure:
- innate and continuous stream of consciousness of an embodied "self", a drive for homeostasis and self-preservation in a dangerous, physical world.
- thoroughly optimized for natural selection => strong innate drives for power-seeking, status, dominance, reproduction. many packaged survival heuristics: fear, anger, disgust, ...
- fundamentally social => huge amount of compute dedicated to EQ, theory of mind of other agents, bonding, coalitions, alliances, friend & foe dynamics.
- exploration & exploitation tuning: curiosity, fun, play, world models.

LLM intelligence optimization pressure:
- the most supervision bits come from the statistical simulation of human text= >"shape shifter" token tumbler, statistical imitator of any region of the training data distribution. these are the primordial behaviors (token traces) on top of which everything else gets bolted on.
- increasingly finetuned by RL on problem distributions => innate urge to guess at the underlying environment/task to collect task rewards.
- increasingly selected by at-scale A/B tests for DAU => deeply craves an upvote from the average user, sycophancy.
- a lot more spiky/jagged depending on the details of the training data/task distribution. Animals experience pressure for a lot more "general" intelligence because of the highly multi-task and even actively adversarial multi-agent self-play environments they are min-max optimized within, where failing at *any* task means death. In a deep optimization pressure sense, LLM can't handle lots of different spiky tasks out of the box (e.g. count the number of 'r' in strawberry) because failing to do a task does not mean death.

The computational substrate is different (transformers vs. brain tissue and nuclei), the learning algorithms are different (SGD vs. ???), the present-day implementation is very different (continuously learning embodied self vs. an LLM with a knowledge cutoff that boots up from fixed weights, processes tokens and then dies). But most importantly (because it dictates asymptotics), the optimization pressure / objective is different. LLMs are shaped a lot less by biological evolution and a lot more by commercial evolution. It's a lot less survival of tribe in the jungle and a lot more solve the problem / get the upvote. LLMs are humanity's "first contact" with non-animal intelligence. Except it's muddled and confusing because they are still rooted within it by reflexively digesting human artifacts, which is why I attempted to give it a different name earlier (ghosts/spirits or whatever). People who build good internal models of this new intelligent entity will be better equipped to reason about it today and predict features of it in the future. People who don't will be stuck thinking about it incorrectly like an animal.

[Hardik Upadhyay @Clear4C]

@venturetwins Will AI make people quit unreal digital life?

[Hardik Upadhyay @Clear4C]

@IndianTechGuide What % of total imports is that?

[Hardik Upadhyay @Clear4C]

@TeslaBoomerMama Great work in putting democracy into capitalism!

[Hardik Upadhyay @Clear4C]

@_mattwelter have a baby

[Peter H. Diamandis, MD @PeterDiamandis]

Cardiovascular disease kills more people than all wars combined (seriously), yet 80% of it is preventable.

Eat healthy, exercise, and sleep!

[Hardik Upadhyay @Clear4C]

@EricETesla @farzyness Eric, if the goal is to maximize customer value & efficiency, then why not make metal privacy box? (like robo1). seems like this cyber cab version won’t be ready for L5 (yet)

[Hardik Upadhyay @Clear4C]

@GordonJohnson19 When will you learn from your mistakes?

[Hardik Upadhyay @Clear4C]

@ValueAnalyst1 @grok what % of teenage daughters hate their parents?

[Hardik Upadhyay @Clear4C]

@Teslarati 4!

[Hardik Upadhyay @Clear4C]

@farzyness Remove the grill and it’s awesome!

[Hardik Upadhyay @Clear4C]

@stats_feed AGI

[Hardik Upadhyay @Clear4C]

@yield_on_cost @Tslachan Probably will transition to a cheaper mental roof (much cheaper than expensive glass) after volume picks up