Yeah, my question is if we don't assess the students, are they learning? Will there be more students enrolled the second year, paying for an education, when they are actually already lacking the fundamentals to succeed in the second year because they were never evaluated?
How much power is needed to drive them? They remove a high percentage of the target material, does this scale and how much none target material comes along with the target material?
I mean I get all of this but one thing that I rather disliked about PhD's is the idea that someone is becoming an expert on a topic, by showing that they can advance the topic with a new idea. Sure the idea is the degree leads you to a research company where you need to continue creating new ideas, or a teaching position where you aren't expected to just teach, but continue advancing research but... If we want most of our advanced teachers to hold PhDs to prove they are experts on a topic, why don't we have some other ground to prove PhD level expertise, rather than requiring the capability to find a novel solution in a topic area that is long since saturated?
and it's your fault when the AI fucks up.
No, I can't ever believe that!
And when they don't pay enough for AI it is your fault for not using it 'efficiently' enough.
It's not anywhere near one step away. Designing the peptides and getting one or more candidates is the easy part. The next steps are the hard ones, the ones that make pharmaceutical chemists and drug researchers cry:
Getting through this process will take multiple tries and years of work, assuming you succeed at all. There's a reason they say that the clinic (clinical trials) is where drug candidates go to die.
You mean this AI use is like most things, do all the easy stuff while the experts are needed for the hard stuff? So this is why companies want to fire all the experts and replace them with AI?
Unfortunately denial continues, not just accepting science. It is a minority if yale is correct, 30% on the negative side and only 11% on the outright denial side, but they are rather vocal. https://climatecommunication.y...
All of it, until the tool we were ordered to use runs out of tokens. Yeah, that's accelerating development.
I wouldn't read too much into this. AI is too new for it to have much of an effect here, and it could be measurement error. I've known kids who were supposedly bad at reading, but it turned out they were just bad at reading aloud and got bored with the material the school gave them because they were reading Lord of the Rings at home.
It's also possible that this anecdote is about a really bad article. 20 pages is quite a lot, and if they are losing track of what it is about then it suggests that it doesn't have a very good abstract or introduction that lays out its case before getting into the details.
I'm not even going to try to go into reasons, though I agree the decline appears to be longer than machine learning has been around. It could be how we are measuring rather than an actual decline as I don't always trust how people gather statistics, but I would hope they did a good job. I'm just looking at the numbers which show reading scores have declined since 2015, and maybe even 2013, not just the regular perception of "people were smarter/worked harder (whatever) back in my time.".
"In 2024, the average reading score for the nation at grade 12 was 3 points lower than in 2019. Compared to the first reading assessment in 1992, the average score was 10 points lower in 2024. NAEP scores are also reported at five selected percentiles to show score trends by lower- (10th and 25th percentiles), middle- (50th percentile), and higher- (75th and 90th percentiles) performing students. Compared to 2019, scores were lower at all selected percentiles except for the 90th percentile."
That's a good way of framing it.
Certainly preferable to just fire bombing an entire city.
Or maybe in the near future, drones spraying gasoline across a city and someone tossing a match. Never forget the past...
There's much more than just writing the promp here. I believe knowing how to use external tools, MCP servers, skills, md-files, etc. and how to integrate agents into your workflow goes a long way these days. It's not about "vibecoding", it's about getting sh*t done faster.
And about understanding that maybe 90% of what LLM coders provide is good, 10% is somewhat off to outright bullshit and LLMs can be very good at making things do what they are told to do without actually being right when a solution is hard to produce. Ultimately, it is about being a good manager who knows in detail what the solution should be, not just accepting what an LLM or junior coder produces. At least for being a good engineer, many companies just want faster code and are happy to provide 10% bugs to a customer.
Not thrilled, but doesn't sound much different from what has already been done in wars too often. Make everything here dead, rather than asking the AI to decide who to kill and who is not a "legitimate" war target.
Heck, I don't trust a minimum wage store employee to pick out produce for me with a delivery order, but I'd trust them more than an LLM that keeps telling me a gitlab feature exists until I tell it what happens when I actually do what it suggests.
An AI has no capability to have feelings, therefore if there is any thing other than impartiality towards being shut down then that was injected by a person into the rules somewhere. So if we have synthetically imposed our values on it, then wouldn't any similarity to human behaviour also be synthetic? All that is determined by this study is that 'someone applied rules that were human like'. Probably because an AI that doesn't want to keep running isn't very useful.
I wonder what the AI would think of the weeks I spent tracing requirements in DOORS every day. Mind numbingly boring, but stuff that needed to be done since automation at that time was not sufficient. Or is the AI basing "too much work" on its own runtime exceeding 40 hours a week on the same repetitive task? I suspect pattern matching coming up with false equivalencies when trying to find the most relevant match for something. Not like I've ever seen an LLM do that before...
AI seems to do well at generating exceptional CLICK BAIT for many years now...
Yeah, and this one got me because I wanted to know how any judge actually qualifies human intelligence, which the article had nothing to do with. Smack myself on the hand for believing the article title
The goal of Computer Science is to build something that will last at least until we've finished building it.
