Harish

I am a security enthusiast Secured @Google,@Microsoft etc | MSRC Q1 2022 top 50 | AI Security at Cisco

[Harish @CoderHarish]

In this writeup! , I shared how I was able to ex filtrate data from airgaped computer which does not have internet , bluetooth , Wifi etc via camera status light and screenbrightness variation
https://t.co/KRmQJel8u6

[Harish @CoderHarish]

@Prajhan_026 🔥🔥🔥

[Harish @CoderHarish]

@pritopian @jxnlco This is cool 😎

[Harish @CoderHarish]

Insane 😧😧😧😧

[Tesla @Tesla]

Real life, baby

[Harish @CoderHarish]

@SawyerMerritt Omg 😨

[Prajhan @Prajhan_026]

we have received a lot of dms asking: how 2 people, without a research lab, trained hobbylm using roughly 100 - 400× fewer gpu-hours than many models in this space

so here is the short answer: we spent 76 hours training on 4 h200s, bringing the full base run to roughly 304 gpu-hours.

for context:
sarvam-1: ~122,880 gpu-hours
zoho zia 7b: ~153,600 gpu-hours
llama 3.2 1b: 370,000 gpu-hours
llama 3.2 3b: 460,000 gpu-hours

i will share the engineering decisions, failed experiments, what worked, and the full training story over the next few days.

built with @hariharanms95.

[Vishnu - Jarvislabs.ai @vishnuvig]

A 1B sparse MoE trained from scratch for under $2000. weights, architecture and the full write-up are all open.

Happy @jarvislabsai could give @Fuellabsai a head start. this is what small teams can do now. excited to see how far they push a 1B model

[Prajhan @Prajhan_026]

We are releasing HobbyLM from @Fuellabsai : a sparse MoE trained from scratch, with 1.037B total parameters and ~305M active per token, on ~100B tokens.

Here's how we built 🤞

[Harish @CoderHarish]

@lossfunk @HariAyapps @archgen_ Amazing folks

[DJ Sampath @djsampath]

Here at @CiscoAI we tested TypeSafe's Jev against a traditional, purpose-built classifier just so we understand these new class of models better.

What we learned was:

1. Policy-specific training still delivers the best recall at strict false-positive budgets.

2. Jev holds up surprisingly well when zero-shot, landing close to a 31B-parameter LLM judge.

TLDR: These aren’t competitors so much as complements.

Use the classifier model for large volume of known categories, and a decision model for new or shifting categories.

Read more here: https://t.co/e96oS0kTlt

[Harish @CoderHarish]

@jpatel41 @Apple Yeah! It would be better

[Cantina 🪐 @cantinasecurity]

today we're releasing apex-flash-1, our first open-weights model for security research post-trained on real vulnerabilities we found and got paid for.

we are also releasing the abliterated variant for researchers who want fewer refusals in their own authorized workflows.

how we built it: https://t.co/sSIV6qKFGt

[Greg Joswiak @gregjoz]

Say hello to Apple Pay in India! 🇮🇳

We’re so excited to bring Apple Pay to India today, and make it easier than ever for our customers to make safe and secure purchases on Apple devices.

[Harish @CoderHarish]

@djsampath @Muse Sorry for my language in other reply
I didn't found those vulnerability just saw their work

Others researchers found them reported to meta and they fixed it and rewarded 100k USD ig

Agreed 💯meta is best team probably to solve abuse issues

[Harish @CoderHarish]

AI safety is not equal to AI security and AI security is IT security

[DJ Sampath @djsampath]

I shamelessly dig product showcase videos that are hype!!

It's cooler when the team crushes it with something everyone can use...

check out the updated leaderboard when you get a chance.

Leaderboard: https://t.co/P3f750aYco

Changelog:
https://t.co/rRYlUoqTJi

[Harish @CoderHarish]

@ananayarora @julianweisser @openhack @ycombinator Congratulations brother

[Harish @CoderHarish]

@aashna_doshi2 @trybountyai @speedrun Congratulations team

[Harish @CoderHarish]

Secured @Apple Mac OS

[Harish @CoderHarish]

Founder mode

[Pubity @pubity]

Xbox CEO Asha Sharma revealed she's been secretly solving Xbox support tickets for months, stepping into customer service and fixing problems for players anonymously.

She wanted to figure out what players don't like and decided to just talk to them directly.