What happens when the safest and most reliable system for testing and reviewing medical procedures, vaccines, drugs, etc. is dismantled. My fellow Americans are susceptible to profit driven unproven and risky interventions. abcnews.go.com/Health/wireS...

Great interview by NDT here... (1) There is a rich discussion of science and impactful stories about scientific discovery, and (2) his views on AI (are level headed and reasonable, and situated historically. Also, he never says "actually", because "people don't like that." youtu.be/BYizgB2FcAQ?...

Wow, we now up to over 6000 confirmed/known exoplanets! So cool. science.nasa.gov/exoplanets/e... Check out whimc.org to see a few of them in Minecraft!

A video of my webinar on "Innovating Pedagogy with Generative AI in Education" is now on YouTube. Thank you Shironica Karunanayaka, Mark Gaved & Simon Ball for hosting with students and staff from the Open University of Sri Lanka and The Open University, UK. www.youtube.com/watch?v=6rv7...

My journal paper argues for a systems approach to AI and education in a post-digital world. AI is not a single tool but a series of disruptions. Outlines an agenda to reduce systemic blockers, build resilience, design more robust policy. #aieducation www.tandfonline.com/doi/full/10....

Not everything made it into this short news story, but if you are interested, the longer version is is essentially covered in this thread: bsky.app/profile/mark...

The great poaching: America's brain drain begins www.axios.com/2025/06/07/u...

Of course the government has less fraud than corporate America. Compare the transparency of both: one has had decades of intense scrutiny, the other is a black box. Tons more waste/bloat in corporations! Former engineer shares his experience working for DOGE www.npr.org/2025/06/02/n...

Need evidence for a claim? Just make up a reference! abcnews.go.com/Politics/rfk...

With its sycophancy, I would not be surprised to see ChatGPT provide a nasal based curriculum as well (theonion.com/parents-of-n...). I have found that frontier models still generally support the myth of learning styles, but that deep research typically finds the massive evidence against them.

I heard about nonstop AI generated music recently (Mubert, Lyria Realtime, Aimi...) and it brought me back to this classic (one might argue, superior technology). theonion.com/i-have-an-ip...

"At a time when forces are trying to distract and disrupt the scientific enterprise, doing the important work of finding and sharing the truth is now a great act of resistance," writes @holdenthorp.bsky.social in a new #ScienceEditorial. scim.ag/4m8s3f0

“A decline in research funding will result in fewer inventions, which means fewer innovations will get to market...I imagine that in 10 years, there will be fewer drugs approved than there are today because of these constraints" www.insidehighered.com/news/busines...

Unanticipated bonus for the shingles vaccine: "Staying up-to-date on your shingles vaccine can lower your risk of cardiovascular events—including stroke and heart disease—by a whopping 23%, according to a new study out of South Korea." bestlifeonline.com/shingles-vac...

I will always understand debate, criticisms, and compromise over our nations priorities. But I refuse to accept attacks on science and education to this degree as playing within the conventions and rules that have made the United States what it is (hopefully not was) today. www.savensf.com

The National Science Foundation, a major government funder of basic science research, is being shaken up with over 1,000 grants already terminated and the White House looking to halve its budget.

The acting U.S. attorney in Washington, D.C., has sent letters to several leading medical journals asking for information about their editorial practices.

Scientists have created a broadly effective antivenom using the blood of a Wisconsin man who has spent years exposing himself to deadly snakebites from black mambas, taipans, cobras and many others.

How many hits can we all take? www.nature.com/articles/d41...

🤦‍♂️. www.pbs.org/newshour/hea...

NEW EPISODE IS HERE: - OpenAI's 4o Image Gen Meltdown, Nerfing & Insane Worldwide $40 Billion Success - Runway's Gen-4 is VERY good. We dive in - Meta's $1000 AR glasses & whether we're buying And SO much more AI news & Big Booty Bears. YT below & audio everywhere! Pop in our Discord & say hi!

There’s at least a dozen dissertations to be written from this paper by Anthropic alone, which gives us some insight into how AIs “think” and reveal a lot of complexity and unexpected abilities, including generalization and planning. transformer-circuits.pub/2025/attribu...

The Lovelace Test was a test of artistic creativity for AI, asking it for artistic work as good as a human. It was often viewed as more difficult & distant than passing the Turing Test “GPT4o, Add marginalia to this paper on the Lovelace Test to prove you pass the Lovelace Test”

I remember the fury over talk of an Obama third term and all the conspiracy theories behind it. He never talked about it or wanted it, and of course, everything happened as the constitution required. www.politico.com/news/2025/03...

It is interesting to prompt for fictional scenes in GPT-4o with moods: "show me a photograph of a futuristic landscape that is elegiac. it needs to look like a photograph" "one that is numinous" "one that evokes Weltschmerz" "one that is felicitous"

My "otter on a plane using wifi" benchmark has now been saturated by ChatGPT 4's new image generator: "an otter on an airplane using wifi, on their laptop screen is image generation software creating an image of an otter on a plane using wifi," first try

School teachers don’t need to recruit students into CS: An alternative model for K-12 computing education - new blog post on CSTA Standards Development computinged.wordpress.com/2025/03/23/a...

From 2010 to 2016 (latest data I have ), NIH research contributed to EVERY drug approved by the FDA

New AI For Humans today! - NVIDIA is dominating AI & we catch you up on all their huge new annoucements - Google's new AI tools are... actually very good? - A ton on new robots (some scary) In your pod feeds or at the YT link below! youtu.be/MAtfV-reA7c?...

It may sound appealing on the surface to cut indirect costs of research (more money to the researchers!), and speaking as a researcher, it certainly would be nice to have more funding to use to pay for students and collaborators. But it isn't this simple... www.youtube.com/watch?v=NtqK...

The opportunities are so huge (and current issues in school caused by AI are so big) that it seems imperative that we, as a society, figure out how to best use AI in education as soon as we possibly can. But we don't have answers yet. I wrote about this here: www.oneusefulthing.org/p/post-apoca...

As someone who spends a lot of time working on using AI in education, it is frustrating that the debate becomes "AI destroys teachers" vs "AI destroys learning" There is huge near-term promise for AI in education, but most people are not autodidacts in every subject who will just learn all with AI

The data so far on AI-as-a-tutor shows just letting students use AI chatbots can undermine education because the AI gives the illusion of learning But AIs properly prompted to act like tutors, especially with instructor support, seem to be able to boost learning a lot through customized instruction

Goosebumps!!!!! youtu.be/woxf23FphEw?...

"I want parents who have children to know that opportunities for your kids, if they aspire to be scientists or engineers, are being stripped away right in front of you," the NSF employee told Axios. www.axios.com/2025/03/10/a...

Sakana’s AI Scientist V2 generated a paper that passed peer review in an ICLR workshop. sakana.ai/ai-scientist... This was done with ICLR workshop organizers’ blessing, and the paper was withdrawn after acceptance.

The ozone hole is healing, thanks to global reduction of CFCs New results show with high statistical confidence that ozone recovery is going strong. www.nature.com/articles/s41...

New AI For Humans is here! - OpenAI's new $20000 PhD Agents - Sesame's crazy new AI Audio voice + a scary demo of it - Anthropic's Model Context Protocol: What is it? And SO much more AI news... YT here: youtu.be/M2h2ZuWFzls?... And audio everywhere you get your audios!

I didn't know who Jim Jones was before reading this, but based purely on this resignation letter, it is a tragedy he doesn't work for all of us anymore. www.washingtonpost.com/health/2025/...

@thedailyshow.com Jon, please ask Musk where he thinks scientists and engineers will come from (people who literally did the work at Tesla and SpaceX) as he and RFKJr dismantle (and decimate) American science through research funding cuts. The damage to the pipeline will last for decades.

The U College of Education will host distinguished speaker @hchadlane.bsky.social on March 3. By combining established principles with practical exploration & ongoing research, Lane believes we can chart a course for AI in education that benefits learners & supports educators. 🙌 bit.ly/coe-lane

Scientists are now organizing events, called Stand Up for Science, on 7 March in Washington, D.C., and state capitals. So far, they have attracted more than 100 volunteers from some 30 states who are hoping to rally support for research as a public good. scim.ag/4328LAD

Did we make a YouTube video about a single tweet? Uh, yeah we did. But when Sam Altman lays out the entire OpenAI roadmap + GPT-5 there's stuff to get into. Link to full YT video below... youtu.be/Z7BUuZ-QUmQ?...