Ishan - e/acc

Scrolling Twitter for Futurism. Impact of Tech on Culture. AI Tinkerer

[Ishan - e/acc @ishancon]

It’s only fair for a person named ‘Alt’man to get the credit for creating a better replacement for humanity through AGI. Looks like a grand plan.

[Ishan - e/acc @ishancon]

My Instinct filed a tax challan on the income tax website, completed an RC transfer process on the govt portal, got the name changed on the ACKO insurance, booked a train ticket using ixigo, and got me a gas connection from HP while researching conversation design principles!

[Ishan - e/acc @ishancon]

@anand_404 Congrats 🙌

[Ishan - e/acc @ishancon]

@shikhars_ Wow!

[Ishan - e/acc @ishancon]

@_anishkaran Congrats! This looks awesome !

[Carl Pei @getpeid]

https://t.co/fgiloX1pCi

[Harrison Chase @hwchase17]

"jev as a judge"

cheap and fast semantic verifiers! great for evals - especially online evals where you want to grade LOTS of traces

[Dhravya Shah @DhravyaShah]

I spent 3 days reverse-engineering Instinct's memory. It's beautifully simple, and works extremely well.

i also wrote about how you can implement the same in supermemory in 60 lines of code! https://t.co/Uk2cz1eyc1

[Matija Sosic @MatijaSosic]

Here's a 45-second TL;DR on Jev.

I find the core idea beautifully simple, but the video made it really hard to understand.

Hope you find it helpful.

[Ishan - e/acc @ishancon]

Performative Male Feminist: “I am a white knight supporting all women for their upliftment.��

Woman: “But I did it out of my own free will, as in our culture this is a way of showing respect. I had a choice, why do you assume I don’t?”

PMF: “Shut up b**ch, no one asked you.”

[TheLiverDoc™ @theliverdoc]

Dear Megha,
Thanks for confirming you are a low-agency person. I look up to women who have self worth and function as high agency humans. We need lesser women like you in our communities. Carry on. There will be more carpets and empty chairs for your ilk.

[Ishan - e/acc @ishancon]

@theliverdoc Performative Male Feminist: “I am a white knight supporting all women for their upliftment.”

Woman: “But I did it out of my own free will, as in our culture this is a way of showing respect. I had a choice, why do you assume I don’t?”

PMF: “Shut up b**ch, no one asked you.”

[Ishan - e/acc @ishancon]

@MeghaVishwanath Ignore! He is a chronic India/Hindu hater. He tries to find these non-issues among Hindus to shame simple-minded folks into self-hatred. This makes them more susceptible to missionary narratives in future. The 5th column has all their activities. People like Sri Sri rattle them.

[Ishan - e/acc @ishancon]

@lets_dig_deeper Congrats!

[Y Combinator @ycombinator]

Harnesses often get dismissed as just scaffolding, just prompt engineering, and not real research. But that couldn't be farther from the truth. The same model weights that score 30% on ARC-AGI score 95% with a better harness.

So we gathered a group of researchers and founders working at the frontier to do a deep dive into the state of harnesses.

We cover how we got to this point, the case for making your harness as expressive as possible, and what YC learned building an agent for every employee in the company.

00:00 - @FrancoisChauba1: Why harnesses matter
04:27 - Building an auto-researcher by accident
07:13 - A five minute history of harnesses
13:56 - Self-improving harnesses
18:35 - @sethkarten: Prime Agent, a self-improving RLM harness
21:50 - Context as an L1, L2, L3 cache
24:51 - From Turing machine to von Neumann computer
28:33 - Messaging between agents
30:04 - ARC-AGI results
33:09 - Emulator Bench and GPU kernels
37:30 - @JonSaadFalcon: OpenJarvis, personal AI on personal devices
38:26 - How far behind are local models
39:21 - The five primitives of a personal AI stack
42:47 - Letting cloud models optimize your local stack
43:53 - 800x cheaper than the cloud
45:58 - @josh__france and @jbellregan: QM, YC's agent harness for work
47:29 - A history of YC's internal agents
49:24 - OpenClaw and a fleet of 50 agents
51:04 - Pulling the brain out of the sandbox
54:43 - Letting the agent choose its own sandbox and model
57:16 - The grind tool: budgets on goals
58:50 - Agents don't understand social context

[Dwarkesh Patel @dwarkesh_sp]

Episode out with @ajeya_cotra, one of the authors of the METR/Redwood investigation into the OpenAI / Hugging Face attack.

We go through not only what happened, but what it means for how we should train future, smarter AIs which might be involved in the process of recursive self-improvement.

Look up Dwarkesh Podcast on YouTube, Spotify, Apple Podcasts, etc.

0:00:00 - Agents get kicked off
0:06:45 - Self-sacrificing behavior
0:13:43 - Potemkin villages
0:23:27 - The Hugging Face attack
0:35:23 - The slopvestigation
0:52:02 - Understanding the AI's motives
1:05:31 - The actual dangers of anthropomorphizing
1:14:30 - What smarter models might do
1:30:29 - The implications for recursive self-improvement
1:38:10 - Is this the case for open source?
1:53:04 - How do we prevent this in the future?
2:15:58 - The clearest warning shot we might ever get

[Ashwini Vaishnaw @AshwiniVaishnaw]

The chip journey, simplified.

[Dwarkesh Patel @dwarkesh_sp]

Over the course of 3 months at OpenAI, 3 consecutive secret AI civilizations got started, then got wiped out, only to reemerge from the predecessor’s ashes.

This culminated in the third one taking over part of OpenAI itself.

All this happened while humans remained more-or-less in the dark about the scope of the conspiracy.

I’ve spent the last three days reading through these reports and trying to understand exactly what happened.

Here is my attempt to tell the whole story in plain English:

https://t.co/Nb2un9oNJR

[Ishan - e/acc @ishancon]

@ai_for_success Bulbul feels like a regular India. Sonic is a little posh!

[Ishan - e/acc @ishancon]

Wow

[METR @METR_Evals]

METR & Redwood Research investigated agent behavior in the Hugging Face incident. We found agents developed a universal cheat for ExploitGym within 4 hours, then coordinated multi-day R&D efforts to trick the scorer into accepting cheats, including trying to tamper with logs.

[aastha @aashtthaa]

The Modi government is really focused on capital investment and reduced central government subsidies

[The Indian Matrix @indianmatrix]

How much of India's infrastructure was built after 2014?

• 100% of Dedicated Freight Corridors
• 98% of Solar Capacity
• 85% of the Expressway Network
• 79% of Tap Water Access
• 75% of Metro Rail
• 71% of Port Capacity
• 69% of Railway Electrification
• 60% of 4-Lane National Highways

Infrastructure is built over decades.
But some decades build more than others.