Prabhu

AI & Research / something for earth.

[Prabhu @prabhugopal_]

@yadavji_codes To make others unemployed.

[Prabhu @prabhugopal_]

@Venkydotdev @AnthropicAI opus is good to have, sonnet is essential :)

[Prabhu @prabhugopal_]

@Oli_Wales @ycombinator the connections and exposure you gain matters. it answers the question of "worth it ?".

[Prabhu @prabhugopal_]

@Amank1412 also he is one of the author of attention is all you need.

[Prabhu @prabhugopal_]

@sourabhkapure Now the codex and Claude code got a new comp !

[Prabhu @prabhugopal_]

@_milosandric It's about distribution and acquiring the customers for good founders.

[Prabhu @prabhugopal_]

looking at how organic ecosystems naturally scale is the best system design masterclass you can get

[Prabhu @prabhugopal_]

If you want zero routing latency, edge inference is the move.

[Prabhu @prabhugopal_]

@0xrinx Came across junior devs talking like " how you guys coded the application before ai ?" many times.. in many places

[Prabhu @prabhugopal_]

@sri9s hey @grok what do you think ? what roles could emerge ??

[Prabhu @prabhugopal_]

@Brennan_Lup architect

[Prabhu @prabhugopal_]

If you're building something that actually needs to matter in 50 years, you don't rush the baseline architecture

[Prabhu @prabhugopal_]

@icanvardar people pricing spacex on backward looking metrics completely miss the plot. It is a triple threat ecosystem... traditional metrics fail when a company holds a literal launch monopoly and a global telecom engine

[Prabhu @prabhugopal_]

@sri9s Yes 🙃

[Prabhu @prabhugopal_]

@rezoundous Pouring a cup of water into a burning house.

[Prabhu @prabhugopal_]

after agi, folks will go for asi. simple !

[Rohan Paul @rohanpaul_ai]

Beautiful paper from Google DeepMind.

Explains the pathways from AGI to ASI, and why that jump could happen through several routes.

The authors frame the AGI-to-ASI transition around 4 technical pathways:

- continued scaling of compute, model size, data, and test-time inference;

- algorithmic paradigm shifts beyond today’s transformer-based foundation-model stack;

- recursive self-improvement, where AI accelerates AI R&D and improves future systems; and

- multi-agent collective intelligence, where large populations of specialized agents coordinate into a superhuman group agent.

Scaling may work for a while, but it could hit limits in data, compute, energy, or weaker returns from making systems larger.

Recursive improvement is the most uncertain path, because AI could speed up AI research, but that loop may also slow if hard research problems need real-world testing, scarce hardware, or new ideas.

Multi-agent collectives may be the most underappreciated path, because a society of competent digital workers could outperform a brilliant individual model through specialization, speed, and coordination.

The big point is that ASI may not arrive as 1 sudden event, but as a chain of faster changes as AI helps create better AI and stronger scientific tools.

----

Link – arxiv. org/abs/2606.12683

Title: "From AGI to ASI"

[Prabhu @prabhugopal_]

@icanvardar now its gone !

[Prabhu @prabhugopal_]

@icanvardar ig the model is more better in cybersecurity stuff, still i was unable to throttle it with those task.. its getting flagged and switched to opus when i go for security tasks !

[Prabhu @prabhugopal_]

@TTrimoreau Imagine a non tech vibecoded guy handing over the codebase to dev who has ocd

[Prabhu @prabhugopal_]

@rezoundous absolute facts, it's like sometime they lock onto one keyword and everything else just vanishes...