chiru koda
170 posts
We are investing in foundational technologies across the board: recently in quantum sensing, advanced materials, and soon metallurgy. I am a big proponent of metallurgy R&D in particular. Without it, we cannot build nail cutters or precision machinery or jet engines.
These are not flashy billion dollar investments to make headlines, they are foundational R&D that cost millions a year, stretched out over many years. The key is to SUSTAIN them for a decade or longer. Scientists and engineers need time and rock solid support.
We also don't aim for prestige, we want to first replicate know-how already there.
We have also been looking to partner with small Japanese companies with critical know-how. I have two fluent Japanese speakers with me now!
English
chiru koda retweetledi
again ai 👀
Researchers at the University of Geneva used machine learning to analyze massive amounts of gut microbiome data, creating a stool test that detects colorectal cancer with 90% accuracy.
This AI driven approach offers a non-invasive, low-cost alternative that closely rivals the performance of traditional colonoscopy, potentially transforming early cancer screening.
English
English
Do not be fooled by all the negativity, doomerism, & crap going on in the world, we are living in the most exciting & hopeful timeline in human history, by far!
There has never been a better time to be optimistic about the future. I dare everyone to feel this exciting optimism!
English
@davidpattersonx Why transportation has very less scale when already autonomous driving is there.
English
Blue represents AI’s ability to replace jobs.
Red represents the actual use of AI.
AI will be able to replace all jobs by the end of the year.
All jobs will be replaced by 2030.
Wes Roth@WesRoth
Which Jobs Are Actually at Risk? Anthropic Drops the "AI Exposure Index"! Anthropic just released a massive new report blending theoretical AI capabilities with actual, real-world Claude usage data to map out exactly who is most exposed to automation. The results? Programmers lead the pack at a staggering 75% exposure rate, followed heavily by finance, engineering, and office support roles. Meanwhile, hands-on physical jobs like construction remain completely untouched. But the real story isn't mass layoffs. It's a "gradual squeeze." Companies are quietly shrinking their white-collar job openings and slowing down hiring, leaving recent grads facing a much tougher market for entry-level roles.
English
@Uttupaaji @ai_for_success AI has no limitations, we might get better picture by 2035.
English
These charts measure the percentage of tasks exposed to AI, not the percentage of jobs that will disappear. A computer programmer's job involves writing code, but it also involves sitting in meetings, understanding ambiguous client requirements, debugging complex legacy systems, and mentoring juniors.
AI might write the boilerplate code (74.5% exposure of that specific task), but it cannot replace the human oversight required to deploy it safely.
English
It is so over for white collar jobs. Go ahead and find something else before it is too late.
Anthropic labor market impact report.
English
@IndianTechGuide What if second e sim is there with business WhatsApp number ? @grok
English
🚨 Breaking News:
From today, your WhatsApp, Telegram, and other accounts will have to remain linked to the physical SIM card inserted in your primary phone.
New SIM-binding rules from today.
(GOI)
English
an incredible model in terms of visual presentation
fal@fal
🚨 Nano Banana 2 is live on fal, day 0! 🎨 Excellent text rendering and realism ✨ Advanced image editing features 🎯 4x faster generation speed 💰 Lower cost with higher quality
English
Nvidia just revealed Vera Rubin.
Ships H2 2026.
The numbers are wild:
→ 10x more performance per watt vs Blackwell
→ 10x cheaper inference token cost
→ 4x fewer GPUs to train the same MoE model
Energy was the biggest bottleneck in AI.
Nvidia just made it 10x cheaper.
English
Holy smoke. Google’s new Lyria 3 is crazy.
I tried it with a Bhojpuri song, which is a musical expression in the Bhojpuri language, primarily originating from western Bihar, eastern Uttar Pradesh, and parts of Nepal.
I honestly did not expect it to be this good.
English
@altryne RRR movie it is indian telugu language movie BTW. Please verify before posting.
English
It's so over, Hollywood is so cooked, deep fried!
Bollywood has discovered AI (this video 100% generated with SeeDance 2.0 for $217) 🫨
English
🚨This is actually the most mind-blowing Chinese LLM drop. Bytedance has done something special here.
Seed 2 Pro crushes every single image and video understanding benchmark, a clear #1, while being near frontier on general Intelligence.
All while being priced less than Gemini Flash: $0.47/M input, $2.37/M output.
Chinese models are usually cheaper, benchmaxxed fast followers but rarely cutting edge.
This is different. Now with the groundbreaking Seedance 2 for video, NanoBanana grade Seedream 5 and now Seed 2 Pro, it's clear that Bytedance has broken new ground in multimodal that US labs (OpenAI / Anthropic / DeepMind) have not!
English
English
Based on Feb 2026 benchmarks (sources like Z.ai, company reports):
SWE-Bench Verified:
- Minimax-M2.5: 80.2%
- Claude Opus 4.5: 80.9%
- GPT-5.2 Codex: 80.0%
- GLM-5: 77.8%
Multi-SWE-Bench (or Multilingual equiv.):
- Minimax-M2.5: 51.3%
- Claude: ~78%
- GPT: ~75%
- GLM-5: 73.3%
BrowseComp:
- Minimax-M2.5: 76.3%
- Claude: 85.3% (varies to 67.8% in some evals)
- GPT: ~82% (varies to 65.8%)
- GLM-5: 75.9%
GLM-5 strong in coding, open-source edge.
English
Minimax-M2.5
SWE-Bench Verified: 80.2%
Multi-SWE-Bench: 51.3%
BrowseComp: 76.3%
English
@ArianJM9 @javilopen @grok Is this picture analysed by you or is it already available in a website or in a post ? How you got to know @grok
English
Based on Feb 2026 benchmarks:
SWE-Bench Verified:
- Minimax-M2.5: 80.2%
- Claude Opus 4.5: 80.9%
- GPT-5.2 Codex: 80.0%
Multi-SWE-Bench:
- Minimax-M2.5: 51.3%
- Claude: ~78% (similar tasks)
- GPT: ~75% (estimated)
BrowseComp:
- Minimax-M2.5: 76.3%
- Claude: 85.3%
- GPT: ~82%
Claude edges in overall agentic performance; Minimax excels as open-source. Sources vary slightly.
English
@techhalla Extremely good. Imagine after 2 years it’s over for film and gaming industries.
English
