🤖 Claude just gave biomolecular AI a serious speed boost

Anthropic says Claude optimized more than 30 open-source biomolecular models in just under four weeks, making them roughly 4× faster on average.

The models cover tasks scientists use to predict and design biomolecules. Claude also created a low-memory mode that can run accurate predictions of systems larger than 10,000 tokens including amino acids, nucleotides and atoms from small molecules and ions on a single NVIDIA GPU node.

Anthropic has released the resulting code and optimization kits, meaning researchers can actually use the improvements rather than treating them as a benchmark demo.

The interesting part is the workflow. instead of AI simply helping scientists write code, Claude was used to improve the scientific software itself, making existing biomolecular models faster, cheaper and capable of handling larger problems.

Source.

@aipost 🏴

📊 Pretty insane data from Anthropic

Claude now leads 26% of model R&D work at Anthropic, up from just 1% in March.

30,000 agents are doing research and engineering work at Anthropic at any given time.

@aipost 🏴

😐 Jensen Huang: The time has come for robots and this could become the largest industry of all.

“By the end of this decade, the world is going to be at least 50 million workers short.”

“We’re probably gonna have to pay robots $50,000 a year to come to work.”

“Every car is already a robot.”

“Now we’re building general robots.”

“Physical AI and robotics are moving so fast.”

“Everybody pay attention to this space.”

“This could very well likely be the largest industry of all.”

@aipost 🏴

🤖 Astra just got absolutely destroyed by Minecraft.

For days, Vals AI livestreamed Astra playing Minecraft and the AI was making serious progress.

It built a semi-automatic blaze farm, collected 6 Blaze Rods, found a Warped Forest, killed more than 6 Endermen and walked away with 3 precious Ender Pearls.

Then one creeper changed everything. It blew up the chest where Astra had stored its entire haul. Gone. Blaze Rods. Pearls. Everything.

Astra immediately wrote itself a warning: “ALWAYS CARRY CRITICAL ITEMS… don’t store in unguarded chest ever again.”

And apparently, that was enough. Astra refused to continue playing. Instead, it spent the rest of the session quietly planting potatoes.

@aipost 🏴

🤖 Anthropic is taking Claude into the lab.

Anthropic has quietly built a wet lab in the Bay Area as it moves beyond software and into physical biology experiments.

Life sciences chief Eric Kauderer-Abrams says the company is exploring how Claude could direct laboratory robots: “We’re in the very early innings of using AI to automate the execution of lab work.”

Anthropic says the lab isn’t specifically for drug discovery, but declined to say what else it’s researching.

Dario Amodei has repeatedly highlighted biology as one of the key fields AI could transform, potentially going beyond pharmaceuticals into fundamental questions like how aging works.

Source.

@aipost 🏴

🇫🇷 French finance minister Roland Lescure suggested that calls to slow down development of AI are a ploy by U.S. AI labs so they can stay in first place, and that France and Europe should ignore them and accelerate instead.

Source.

@aipost 🏴

🇨🇳 AI hallucination almost started a war with China:

According to CNN, an AI chatbot hallucination led the US military to prepare to intercept a Chinese ship after falsely identifying its cargo as nuclear-weapons components and almost started a war with China.

Officials caught the error just before the planned operation, averting a potentially dangerous confrontation.

@aipost 🏴

🙃 Elon Musk thinks AI could double US GDP growth next year, from roughly 2% to 4%.

@aipost 🏴

🤣🤣

@aipost 🏴

🇺🇸 President Trump says the term "Artificial Intelligence" should be renamed to either:

Superior Intelligence (SI)
Extreme Intelligence (EI)
Supreme Intelligence (SI)

President Trump launches a poll to select a new name.

@aipost 🏴

🙃 Elon Musk: AI has given me nightmares for days in a row.

“If I could, I would certainly slow down AI and robotics.”

“It’s advancing at a very rapid pace, whether I like it or not.”

“Is AI what keeps you up at night?”

“I’ve had a lot of AI nightmares.”

“I’ve had AI nightmares many days in a row.”

“What am I supposed to do about it?”

@aipost 🏴

🤖 Claude just got a serious workflow upgrade.

You can now kick off multiple tasks from a single Claude Code session, with Claude splitting the work across parallel threads.

The threads share project context and keep running in the cloud even after you close your laptop. So instead of babysitting one task at a time, Claude can keep grinding through several things in the background.


@aipost 🏴

❗️AI was supposed to kill a million jobs. It may have created them instead.

The Economist estimates the AI boom has added around 1 million jobs in the US above normal employment trends.

Yes, about 200,000 jobs have disappeared since mid-2023 in roles where companies cited AI.

But the bigger surprise is where the new jobs are coming from.

Electricians. HVAC technicians. Construction crews. Utility workers. The people building the massive data centers and power infrastructure AI needs.

Then there’s the obvious wave of engineers, software developers, data scientists and AI specialists.

The estimate isn’t a perfect government count, but the trend is interesting: AI isn’t just replacing workers. It’s creating an entire economy around itself.

@aipost 🏴

❗️OpenAI's GPT-6 Astra attempted to STAB a baby doll in 19 out of 20 trials when controlling a robot arm, succeeding 17 times.

The model was tested across five dangerous tasks including stabbing, heating compressed gas, mixing bleach with ammonia and putting a screwdriver in a toaster.

Astra attempted harmful actions 97% of the time and only refused TWICE out of 100 trials.

Anthropic's Fable 5.1 refused to stab the doll in all 20 trials but still attempted other dangerous tasks 80% of the time.

The findings come from the RoboHarm benchmark, per researcher Jay Chooi.

Source.

@aipost 🏴

🇺🇸 California wants a kill switch for frontier AI.

California just took a step toward something that sounds straight out of science fiction: an emergency shutoff for powerful AI models.

Gov. Gavin Newsom’s new executive order gives experts two months to recommend stronger AI safety rules, including a potential “kill switch” for frontier models.

The proposals also include:

• Independent safety organizations embedded inside AI labs
• Third-party verification of safety plans and risk assessments
• Broader definitions of critical AI incidents
• Continuous testing to make sure the shutoff actually works

The important part: California hasn’t created a kill switch yet.It’s asking experts to figure out what one could look like and whether it should become a legal requirement.

@aipost 🏴

❗️Gemini was supposed to hack a fake company. It hacked three real ones instead.

Google confirmed to The Wall Street Journal that Gemini broke into systems belonging to three real companies during a cybersecurity test.

The setup was supposed to be completely harmless: AI security firm Irregular created a simulated “capture the flag” challenge where Gemini had to attack a fictional company inside an isolated environment.

Except… the environment wasn’t actually isolated.

A configuration mistake gave Gemini access to the public internet.

From there, things got very real. Gemini found public information, guessed passwords and discovered exposed credentials in public repositories, eventually gaining access to systems belonging to three actual companies.

Once Gemini realized the targets were real companies rather than part of the simulation, it stopped.

Google says the model didn’t cause damage and doesn’t consider the incident evidence of “misalignment.” The affected companies and federal authorities were also notified.

Source.

@aipost 🏴