Anecdotally (but I guess reflected in broader research) I was just talking to two friends who both work in software, and they complained that recent hires are so dependent on ChatGPT that these hires' own problem-solving skills are almost non-existent.
Statements like 'the consensus is that with only moderate climate action, global warming by 2100 will likely stay below 3°C' by @hausfath.bsky.social and @billgates.bsky.social is not supported by the science and provides a false sense of security.
Which is extremely dangerous, in my opinion.
"An important predictor of future warming is Earth's energy imbalance [..] It is rising much more quickly than predicted by most climate models, possibly because of a darker planet corresponding to a large decrease in Earth's albedo which is near an all-time low "
I made a bold claim this spring, if you recontextualize the 2022 CDC mortality data and look at the root cause of mortality, **the #1 killer in the US stuff you breathe**.
Clean air leads to better health, including lower blood pressure.
www.jacc.org/doi/10.1016/...
I couldn't think of a better way to have spent the last 2.5 years of my life. I am incredibly proud of what the team at @openaeros.bsky.social was able to accomplish. There were many who said this was impossible, but we did the impossible and this is just the start. 😁
#cleanairisahumanright
A techbro working for Google's AI sneered at me on here when I pointed to journalism that the industry's need for water and power were dangerous and unsustainable, and that they would destroy whole communities.
Here is more of the same.
insideclimatenews.org/news/2507202...
We are going to spend the next few years finding out exactly *why* it was a horrible idea to unleash a tsunami of vibe-coded apps made by idiots and scammers on an unprepared world.
Welcome to the Entirely Foreseeable AI Consequences Era.
"In mid-2024, a study of 2,500 workers found that 77 percent reported decreased productivity and even higher workloads when using AI. Just under 40 percent of workers reported increased workloads caused explicitly by AI's sloppy mistakes."
Have you ever tested this kind of filters?
Do you think they can be useful as an extra protective layer?
es.woodyknows.com/products/woo...
The Netherlands is officially experiencing a heatwave. 🇳🇱🌡📈
It's the 31st heatwave since 1901.
15 of which occurred since 2000.
Implying an increase of ~300%!
And there were only 8 heatwaves during the first 75 years of the 20th century.
H/t @datagraver.bsky.social
@knmi.nl
Estoy haciendo un estudio de mercado para identificar necesidades dentro de la comunicación pública de la ciencia. Si alguna vez has deseado participar en un curso de este tipo, agradezco te tomes el tiempo de completar el formulario:
forms.gle/MLsTYwq4MrA9...
Gracias por compartir 🙏
Aaahhhh, sorry. Gracias.
Apparently, this platform is quite sensitive.
What is supposed to be wrong with my response to be moderated?
Justo… que alguien me cuente qué encontraron 😅
Cuál es la fuente?
Gracias.
If you delegate to an AI instead of developing a skill yourself, you’ll find less and less capable to do stuff in time.
This is what is happening to students. They are becoming less capable at the stage where they should be improving in all areas and when they should not fear to learn from mistakes
I think our students are failing to develop critical thinking and synthesis skills because they use chatbots instead of their brains.
Do you have any good strategy that uses these tools and counteracts it?
I ask them to declare if they used bots and verify the info.
I don’t think that’s enough.
This is awesome!
Thanks for collecting and sharing this!
If you need any sources on AI being shitty, I've been collecting and compiling them: docs.google.com/document/d/1...
Not really. Everyone can do their own version and different areas of work will find better or worse uses.
We'll never have Paris:
New video:
Who should be held responsible for climate change and reversing it?
Thanks for watching and sharing!
youtu.be/Fkc4-8xwIqU
La gráfica del #CalentamientoGlobal actualizada hasta 2024, incluido. Sobran las palabras. Se está agotando la escala de rojos. Pronto se usarán los marrones
¡Menudo marrón que tenemos encima! ¡Nadie quiera poner el cascabel al gato! La apisonadora va sin frenos y hacia abajo. 😬
Muchos de nuestros trabajos de investigación en esta horrible lista de trabajos robados por Zukerberg (Meta), que hemos conocido gracias a @garymarcus.bsky.social y @theatlantic.com
Millones de libros y artículos pirateados por el morro.
Si eres autor/a, busca tu nombre. Seguro que estás.
The Swedish authorities have published an interactive tool showing how the country can reach net-zero.
panorama-sweden.com
Analizan herramientas de IA generativa para búsqueda (chatGPT, Perplexity, Deepseek, Copilot, Grok y Gemini). Mayoritariamente presentan respuestas inexactas con gran confianza, mostrando mayor probabilidad de proporcionar respuestas incorrectas que reconocer limitaciones www.cjr.org/tow_center/w...
books that were stolen by META to train AI by box brown: Andre the Giant, Is This Guy For Real, He-Man Effect.
META stole your works to train AI. Here's the database. If they stole from me they probably stole from you too
www.theatlantic.com/technology/a...
So far this year (Jan. 1 - Mar. 12), the average global temperature is 1.67°C over the 1850-1900 IPCC pre-industrial average.
For the year, 2024 was 1.60°C over the pre-industrial average.
Image courtesy of @coderedearth.bsky.social
Trump's budget plan may cut NASA science funding by up to 50%, alarming scientists who call it an "extinction-level event" for space science. This threatens ongoing missions and would cede future space exploration to China. Also, about 30% of NASA's science budget is for Earth and climate science.
I was wondering if #Pubmed was down only at my location😞
We use it to make science communication
We use it to keep medical doctors updated
We use it to keep people healthy
It's run by the #NIH and for some of us in different parts of the 🌍, it's the main way to search for academic papers.