post-training image/video models @meta
San Francisco
- Tweets 10,451
- Following 911
- Followers 989
- Likes 41,604
Mathieu Tuli retweeted
Yesterday, we released a cyberdefense strategy for superintelligence: Secure Acceleration.
Here are our calls to action:
1. We must build a competitive American open-source model and invent ways to detect AI sabotage
2. Containment should become a national security priority and we must develop capabilities to shut down untethered models
3. We must secure frontier model weights and RSI without impacting AI development
Secure the path to superintelligence.
secureacceleration.com
6
5
35
13,714
If you wanted to create you would create.
AI making it easier does not make you creative.
Cheaper sure, not "more creative".
That narrative does a lot of damage to the space.
Prompting, like prompting Google, doesn't make you creative.
Create because you want to.
39
Great read, superintelligence doesn't matter if you can't secure it
We are approaching cyber-superintelligence.
But our systems aren't ready. We must secure our model weights and infrastructure against three new threats: sabotage, escape, and theft.
There is a way forward.
Secure Acceleration:
A Cyberdefense Strategy for Superintelligence
2
217
maybe we shouldn't tell models they'll die if they don't finish the task. that might help alignment.
1
53
Mathieu Tuli retweeted
Right-Click Anything: a little demo combining Muse Spark, SAM 3.1 and Muse Image -- now all on @Meta Model API! Muse Spark captions the scene and generates noun-phrases. These are used to prompt SAM 3.1 which predicts instance masks for each object. Then, right-click any object mask as you would with text on any web page! “Look Up” hits Muse Spark web search for real product listings, “Copy” gives you the alpha-matted PNG, “Remove” paints the SAM 3.1 mask into the image and asks Muse Image to inpaint it away. SAM 3.1 also supports video so many more possibilities there too! Have fun combining models across Meta's API platform!
4
11
141
8,545
trailers being a requisite for movie/show watching, cutting good and contextual dialogue/scenes is a sin.
68
Pretty simple suggestion for reducing rage baiting and false statements on X: when you’re caught and community noted, you get a small fine and post-blocked for 3 days, with your post reachability reduced 20%. The more you are caught the more this compounds.
1
123
If you're interested in the specialized model world, and how it helps scale, checkout my talk from buffconf on how we're thinking about this at Meta
piped.video/watch?v=CMd2cwcN…
Training tiny models for special purpose use cases works so incredibly well if you have a great self improving recursive flywheel. Shopify ML team is on fire.
finetuned 0.8b model beats GPT 5.6-sol xhigh in this very specialized task.
1
270
Mathieu Tuli retweeted
we're opening muse connectors to developers!
we have seen so much excitement in the developer community, integrating muse into everything from robots to mood lights and more.
plug your API into muse, then people can use your service just by asking for it. every request runs in a secure VM, and muse will ask before anything consequential.
come build with us! muse.ai/platform
126
93
1,223
277,633