“Very impressed with Gemini 2.5 for coding – it actually tells you when it can’t do what you asked, whereas Sonnet tends to just power through and give you a wrong solution.” / X https://x.com/dzhng/status/1906575275997167857
Taking a responsible path to AGI – Google DeepMind https://deepmind.google/discover/blog/taking-a-responsible-path-to-agi/
An_Approach_to_Technical_AGI_Safety_Apr_2025.pdf https://storage.googleapis.com/deepmind-media/DeepMind.com/Blog/evaluating-potential-cybersecurity-threats-of-advanced-ai/An_Approach_to_Technical_AGI_Safety_Apr_2025.pdf
Alphabet’s AI drug discovery platform Isomorphic Labs raises $600M from Thrive | TechCrunch https://techcrunch.com/2025/03/31/alphabets-ai-drug-discovery-platform-isomorphic-labs-raises-600m-from-thrive/
“Try out Gemini Code: a coding assistant in your terminal powered by Gemini 2.5 Pro. 💜 https://x.com/raizamrtn/status/1906727510601355393
5 new AI education initiatives from Google https://blog.google/outreach-initiatives/education/ai-literacy-day-2025/
“Gemini 2.5 Pro is taking off 🚀🚀🚀 The team is sprinting, TPUs are running hot, and we want to get our most intelligent model into more people’s hands asap. Which is why we decided to roll out Gemini 2.5 Pro (experimental) to all Gemini users, beginning today. Try it at no” / X https://x.com/GeminiApp/status/1906131622736679332
“Gemma 3 can do function calling and is now in Berkeley Function-Calling Leaderboard! 🦍 Leaderboard: https://x.com/osanseviero/status/1907055078776725960
“Gemini 2.5 Pro is really amazing. Scored top (84%) on GPQA Diamond, which is a very difficult multiple-choice questions about biology, chemistry, and physics; human experts only score around 70%. @GeminiApp https://x.com/rohanpaul_ai/status/1907601383315931580
“Gemini 2.5 Pro got 24.4% on the 2025 US AMO (America Mathematical Olympiad)! https://x.com/danielhanchen/status/1907555378067640359
“Gemini 2.5 Pro scores 130 IQ on Mensa Norway https://x.com/mark_k/status/1906106781232931199
“ChatGPT briefly overtook Google as a search term ! https://x.com/BorisMPower/status/1906329961730130239
Google-backed Isomorphic Labs raises $600m to advance AI drug discovery https://finance.yahoo.com/news/google-backed-isomorphic-labs-raises-173041674.html
Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison – Composio https://composio.dev/blog/gemini-2-5-pro-vs-claude-3-7-sonnet-coding-comparison/
“NEW: Google is in advanced talks to rent Nvidia Blackwell chips from CoreWeave, as well as to potentially house its TPUs in Coreweave facilities. The deal highlights the intense customer demand for compute: w/ @aaronpholmes @anissagardizy8: https://x.com/steph_palazzolo/status/1907517483524686129
Google is shipping Gemini models faster than its AI safety reports | TechCrunch https://techcrunch.com/2025/04/03/google-is-shipping-gemini-models-faster-than-its-ai-safety-reports/
“One reason I’m not using Gemini 2.5 Pro is because it doesn’t render math using LaTex like ChatGPT. Google did a good job, but they lose on these little details.” / X https://x.com/hkproj/status/1907766301109403890
“Before 2020, most of the AI field was open and collaborative. For me, that was the key factor that accelerated scientific progress and made the impossible possible—just look at the “T” in ChatGPT, which comes from the Transformer architecture openly shared by Google. Then came” / X https://x.com/ClementDelangue/status/1906854756225630420
Google Gemini shakes up AI leadership, Sissie Hsiao steps down, replaced by Josh Woodward | Semafor https://www.semafor.com/article/04/02/2025/google-gemini-shakes-up-ai-leadership-sissie-hsiao-steps-down-replaced-by-josh-woodward
“Interesting to see the uplift on HMMT This exam is notable because it’s generally considered to be harder than AIME, and it took place after the models’ knowledge cutoff. Gemini’s rate of progress in math is amazing to see, driven by talented researchers like Ankesh!” / X https://x.com/jack_w_rae/status/1906868244654473601
“Google’s new Gemini 2.5 Pro Experimental model has started rolling out to all users, including those on the free tier 2.5 Pro also scored a 130 on Mensa Norway’s IQ test, well surpassing other AI models and the average human’s score of 100 https://x.com/rowancheung/status/1906583578722390160
“It doesn’t feel like Gemini models are ahead. Because Google keeps on doing “exp” models. I can’t take them serious without knowing the price. like when was the last real model release? they haven’t even shipped Gemini 2.0 Pro” / X https://x.com/scaling01/status/1906502465869971507
“some notes: Gemini 2.5 Pro production version with pricing will come “very soon hopefully” “In the more immediate term we want to bring the 2.5 series to more models … Flash is the next on the list” Gemini 2.5 Pro has dynamic thinking but is not yet there where they want it” / X https://x.com/scaling01/status/1906748722572079438
“Let’s recap the releases in open AI this week 🧶 they were few but high quality ones! Multimodal 🖼️ > @Alibaba_Qwen released their new Omni (many input to many output) model Qwen2.5-Omni-7B which beats Gemini 1.5 Pro on OmniBench (OS) > @Alibaba_Qwen released a new iteration https://x.com/mervenoyann/status/1906046978641023458
NotebookLM Discover Sources: Add web research to your notebook https://blog.google/technology/google-labs/notebooklm-discover-sources/
“Being told that you are wrong is actually a game-changing feature of Gemini 2.5 Pro! The real problem with AI isn’t that it makes mistakes, but that it can convince humans that wrong answers are right.” / X https://x.com/rasbt/status/1907618232699109615
Gemini 2.5: Our newest Gemini model with thinking https://blog.google/technology/google-deepmind/gemini-model-thinking-updates-march-2025/
“Google DeepMind has released a new flagship model, Gemini 2.5 Pro. We evaluated it on GPQA Diamond, and found a score of 84%, exactly matching the result reported by Google. This is the best result we have found on this benchmark to date! https://x.com/EpochAIResearch/status/1907519991252992508
“Ideogram 3.0 launched with new text rendering and graphic design capabilities —Creates complex layouts, logos, and typography —Outperforms Google’s Imagen 3, Flux Pro 1.1, and Recraft V3 —Style References to control generations —Available to free users https://x.com/rowancheung/status/1905130195738075507
“Adobe premiere is FINALLY getting media intelligence and semantic search! I’ve wanted this ability to pull up the right clip based on a simple description the moment I tried it in Google photos. Huge when you’re trying to find that *one* b-roll clip. https://x.com/bilawalsidhu/status/1907516479920107689
“I am super excited for AI for scientific innovation, a direction that will certainly grow in the next five years. I think there will be two flavors of it. The first is “deepmind style”, where there is a very specific, important problem to solve (e.g., protein-folding), and you” / X https://x.com/_jasonwei/status/1907532120446349353
“Waymo is a masterpiece. The rendering of your initials on LiDAR. The automatic unlock as you approach the car. The driving starting only after you fasten your seatbelt. The 3D map that reconstructs the world. The greetings. The flawless track record. Only in Silicon Valley. https://x.com/rauchg/status/1907989933731754088
How Google built its Gemini robotics models https://blog.google/products/gemini/how-we-built-gemini-robotics/
“NEW MODEL: GemmaCoder3-12b is a code reasoning model that improves performance on the LiveCodeBench benchmark 11 points over the base model. This makes for a useful code model because: – At 8 bit it runs nicely on 32gb of RAM – Gemma3’s 128k context length is great for large https://x.com/ben_burtenshaw/status/1907074356997423322
