Sarah Cogan
@sarah_cogan
Followers
475
Following
8K
Media
20
Statuses
491
existential risks are bad. I’m tall. SWE @GoogleDeepMind Frontier Safety
San Francisco, CA
Joined March 2016
We just launched Gemini 3.0 Pro this model is a step up in capability across many domains, requiring careful testing Frontier Safety Framework report link below 👇
3
14
90
I hate San Francisco halloween, what do you mean you’re the cat that got hit by a waymo
36
32
859
is it over for me if i'm unironically like this
Met this beautiful Chinese girl yesterday. She was standing outside of my office of all places! She approached me, and suggested we go to impromptu drinks. She was so interested in my work & LLMs. Asking me all these technical questions. She’s so smart. I might be in love.
7
3
261
Just released the 3rd iteration of our Frontier Safety Framework. This update includes a new Harmful Manipulation CCL and more information around our risk assessment processes. https://t.co/5u23eHqjAa
As we build increasingly powerful AI models, we’re committed to responsible development. We’re implementing our latest Frontier Safety Framework – our most comprehensive approach yet for identifying and staying ahead of emerging risks. Find out more → https://t.co/fsjYhxLXm5
0
6
46
"No man steps in the same river twice for he is not the same man and it is not the same river" hits a little different every time I hear it
75
1K
18K
i'm going to become the joker
The reality is she has a lisp, lower voice, and all her interests revolve around having someone be her adventure buddy and follow her on her travels, which is fundamentally more masculine and she exudes more masculine energy. If she was more gentle, soft-spoken, and feminine,
0
0
5
We just shipped Gemini 2.5 Deep Think it doesn't just recall research papers - it fuses ideas across papers in ways I haven't seen before this level of capability demands careful evaluation model card below 👇
38
148
1K
A simple AGI safety technique: AI’s thoughts are in plain English, just read them We know it works, with OK (not perfect) transparency! The risk is fragility: RL training, new architectures, etc threaten transparency Experts from many orgs agree we should try to preserve it:
38
115
452
pro tip: if you ask for birthday presents, not only will you get gifts, your friends will commend you for asking!
0
0
2
As models advance, a key AI safety concern is deceptive alignment / "scheming" – where AI might covertly pursue unintended goals. Our paper "Evaluating Frontier Models for Stealth and Situational Awareness" assesses whether current models can scheme. https://t.co/PrfZcIVuEw
17
45
216
Full transparency, I did not know about this section on pages 278-279 of the OBBB that strips states of the right to make laws or regulate AI for 10 years. I am adamantly OPPOSED to this and it is a violation of state rights and I would have voted NO if I had known this was in
0
0
9
was legit worried this was about me until I remembered I was rehearsing *my* breakup with Gemini Live in the Mission
eating sushi in salesforce park, listening to the woman next to me use chatgpt to rehearse her breakup conversation
1
0
34
Trying to land this plane safely on the Hudson
if you're 25+ and listening to evanescence or creed, wtf are you doing?
15
2K
32K
you would not believe what it took to get this into the dataset for the ai to parrot it (stochastically)
Veo3 artık Türkiye’de. ✨ Gemini’ı sen de dene. Prompt: Bol fıstıklı baklavadan yapılmış bir bilgisayar klavyesi hayal et. Bir kişininin ellerini baklavadan tuşlara basarak dizüstü bilgisayarında yazı yazarken görüyoruz.
22
48
2K