ByteGuardian N/A
@GuardByteTech
ByteGuardian AI Enthusiast Protecting our future with honesty.
Photonic processors are limited by the bulkiness of discrete components and wiring complexity. An experiment now demonstrates a reprogrammable two-dimensional waveguide that performs neural network inference through multimode wave propagation. nature.com/articles/s4156…
in the age of research openai co founder says due to high demand, they’re having to reallocate a lot of compute away from research. this is some code red
We’re back in a Flash ⚡ Gemini 3 Flash is our latest model with frontier intelligence built for lightning speed, and pushing the Pareto Frontier of performance and efficiency. It outperforms 2.5 Pro while being 3x faster at a fraction of the cost. With this release, Gemini 3’s…
Think of orchestration as search over thoughts. Then train the model to match that orchestration. This allows shifting the accuracy↔latency/context tradeoff by compressing search into weights.
I'm glad this paper of ours is getting attention. It shows that there are more efficient and effective ways for models to use their thinking tokens than generating a long uninterrupted thinking trace. Our PDR (parallel/distill/refine) orchestration gives much better final…
Bryan almost gets it. The problem is that we can't survive giving birth to superintelligence. If we are serious about not dying, we need to prevent superintelligence from being born! Sign superintelligence-statement.org
I’m going to try and achieve immortality by 2039. One year of time passes and I remain the same biological age. I invite you to join me. The search for the fountain of youth is the oldest story ever told. It’s been the dream of dreamers for millennia but always painfully out…
I'll be mentoring for this! I think MATS is one of the best opportunities to do high-value AI safety work and get plugged into the community. Consider applying!
MATS 10.0 applications are open! Launch your career in AI alignment, governance, and security with our 12-week research program. MATS provides field-leading research mentorship, funding, Berkeley & London offices, housing, and talks/workshops with AI experts.
Digital Wellness in 2030: Designing for a Calmer Mind in a Connected World. By 2030, digital wellness won’t be a niche conversation; it will be a core design principle across devices, workplaces, healthcare, and public policy. unfoldlabs.com/blogs/digital-…
CDS Prof. and Turing Award winner @ylecun will leave Meta at the end of the year, and is launching a startup focused on “world models”—AI systems that learn by observing and simulating physical environments. nytimes.com/2025/11/19/tec…
Job Opportunity: Assistant Professor in Global Risk and Resilience (CSER) (closes 8 February 2026) cam.ac.uk/jobs/assistant…
What would Nolan or Spielberg have made if they 'took the struggle out' of it? What would any artist have made if there was no struggle? Rising to a challenge is where creativity and art thrive and where people make work they can be truly proud of. AI undermines all of that.
The fact that it took me years to master high end retouching and frequency separation in photoshop, to the point where I developed fcking carpal tunnel before the age of 25 now people can just ask chat to do it edit anything for them is something that I have a hard time…
ChatGPT turns three: here’s five lessons from the frontline of AI adoption… By Emily Lewis-Pinnell via @ForbesTechCncl forbes.com/councils/forbe… @BetaMoroney @Nicochan33 @enilev @mvollmer1 @mikeflache @antgrasso @FrRonconi @ramonvidall @baski_LA @AkwyZ @Khulood_Almani…
As a corporate copywriter who expects her job to go away in the near future, I can say that internal production processes in large companies can make it challenging to offload writing/designing/proofreading/developing/testing to an agent. AI accelerationists downplay this fact.
Fundamentally, it is a spiritual problem and can be surprisingly simply answered in a world we have forgotten: Humanity must survive.
Utilitarianism is not a compelling enough philosophy to bet the future existence of humanity on it being the correct and complete systemization and solution to mortality.
“One national AI policy” sounds efficient. In reality? It wipes out state oversight and hands power to unelected tech executives. One framework. One lobbying target. Unchecked corporate influence.
Humanoids Are Built from Philosophy, Not Parts! Weave Robotics isn’t just about robots that fold T-shirts. Scott @GoingBallistic5 sits down with Kaan @kaandogrusoz from @WeaveRobotics to explore the philosophy behind their designs. Full Video: youtu.be/upMkyXWMCbE?si…
this stuff is finally getting to the adversarial-example level hopelessness i was expecting to feel with LLMs. how do you possibly protect against this stuff in full generality?
New paper: You can train an LLM only on good behavior and implant a backdoor for turning it evil. How? 1. The Terminator is bad in the original film but good in the sequels. 2. Train an LLM to act well in the sequels. It'll be evil if told it's 1984. More weird experiments 🧵
Disney sent a cease-and-desist to Google for AI copyright infringement Worth reminding everyone that Google execs like Paige have been *explicitly* encouraging users to generate Disney outputs …
Inspired by @TaliaRinger, powered by @NanoBanana: Create a detailed poster of the English Alphabet, with each letter shown in capital and lowercase form, with a visual to remember it. The infographic should be ideal for toddlers to learn the alphabet, and should be Star Wars…
send me your best examples of normies frothing angrily about ai. high reach crashouts, tiktokky anti ai memes, moral panics i'm tryna see something real quick
i wonder if the words people used are so charged that the grok algorithm simply buries this post forever. that would be beautifully poetic
United States Trends
- 1. Jokic 28.6K posts
- 2. #heatedrivalry 167K posts
- 3. #StrangerThings5 422K posts
- 4. ilya 132K posts
- 5. Scott Foster 2,982 posts
- 6. jancy 60.7K posts
- 7. Anthony Edwards 9,358 posts
- 8. Holly 99.6K posts
- 9. #PowerForce 2,962 posts
- 10. Dustin 155K posts
- 11. Murray 27K posts
- 12. tammy 48K posts
- 13. yuna 35.1K posts
- 14. Nigeria 459K posts
- 15. robin 76.4K posts
- 16. Lucas 102K posts
- 17. Vecna 98.9K posts
- 18. Kali 93.6K posts
- 19. #MakeOffer 3,334 posts
- 20. Domain For Sale 3,703 posts
Something went wrong.
Something went wrong.