If you give the LLM the professor's prompt and turn in what it produces, you won't learn anything. If you write your own essay (perhaps using an LLM to find sources) and then go back and forth with an LLM to improve it, LLMs may help you learn.
Another WSJ story is "Everyone’s Talking About AI Agents. Barely Anyone Knows What They Are. ‘Agentic’ is the biggest buzzword in Silicon Valley, but tech companies and enterprises lack a common understanding of its meaning, and it’s causing problems"
https://www.wsj.com/articles/everyones-talking-about-ai-agen...
Ambiguity is a feature for propaganda. If people project random hopes
and fears onto buzzwords that gets them talking and serves sales
well. In an imaginary world where people were highly educated and
journalists honest and specific, new advances would be taken-up
quickly by a small group of beneficiaries and everyone else would move
on. The idea of software "agents" is of course nothing new. "Buzz" -
AKA "the warm thrill of confusion and space-cadet glow" appeals
primarily to the idea that "this changes everything!". The smoke and
mirrors machine never sleeps.
I think you're right for a reasoned debate we need to understand each sides view. I don't like the IPR outcome in private health research, and I think the conflation of real-world dollar value with social value in health is a giant mistake. But it would be naieve to think money is not a giant motivating factor in things like the emergence of mRNA solutions, or the semaglutide type drugs.
Novo Nordisk became (briefly) the largest component of the Danish balance of payments for a while there.
So in that sense, I agree with the headline position Reason argues for here. Research will happen, and we won't "fall off a cliff" in terms of outcome. But the other side, the relationship of public health benefits and public goods, and the access to healthy long lives by everyone worldwide has now been tied up in IPR, and will depend on the benificence of the IPR holder, not outcomes of state funded research which become public goods in the wider sense. In time of course, this solves. There is a strong element of belief that "in time" applies to IPR and means we all get free disney in 300 years, hurrah: I don't personally find this very compelling.
I think the idea "money backs winners" is highly risky in health because what emerges is "money backs money making winners" -we don't get 1c per day cures for diabetes, we get $1000 per week cures. Money does not like solving underlying public health problems for a cheap solution: why would you do that to money?
Why not let the plane land, fine the airline, and detain the pilot in the airport until he can be flown back, instead of wasting hours of time of hundreds of people?
Because it's the pilot who makes the choice. Intentionally going to a foreign country without the appropriate documents is a bad idea with unpredictable consequences. Unless the destination country gives legally binding guarantees that nothing bad is going to happen, it's not something a reasonable person is going to do. Especially for a trivial reason such as work.
When hiring, one should strive to avoid racial bias, but for dating I don't think people should be criticized for having a racial "bias", which I would describe using a more neutral word such as "preference".
I grew up in Spain––as a half-ginger––so know all about being a "minority" when it sux to be. But one quickly learns to play the field, and what a fun field it is. +1 on zero tolerance in workspace tho. Merit based or home.
'The word “vibe” has become so ubiquitous that it’s started to lose its meaning. A buzzy new phrase, vibe coding, has gripped the tech industry just a month after it was coined by Andrej Karpathy, a co-founder of OpenAI and venerated programmer who posted on X that it was “where you fully give in to the vibes, embrace exponentials, and forget that the code even exists." In other words, let AI do all the coding for you.
Since then, the phrase has popped up everywhere in tech circles — on social media, forums like Hacker News, executive posts on LinkedIn and in Slack channels populated by startup founders in the Y Combinator accelerator in California. “The term is blowing up,” says Chaz Englander, founder of Model ML, which went through the program last year.
...
It was once rare to see startups with fewer than 20 people bringing in tens of millions of dollars in annual revenue, but now it’s becoming the norm, says Englander. Little wonder that AI coding startups like Stockholm, Sweden-based Lovable have become the hot new thing among venture capital funds. But let’s not get carried away. Businesses tempted to emulate Salesforce should remember that enterprise-grade code still requires plenty of humans. Vibe coding isn’t the answer to your budgetary problems. For one thing, today’s newest startups are leaner not just because they’re using AI in code creation, but also in sales and customer service. “When our sales team arrives in the morning, there are already between 30 and 50 emails in the outbox,” all generated by AI and awaiting human review, Englander says. And while vibe coding is fine for making prototypes and simple pieces of software, you absolutely can’t build enterprise-grade software just using AI. Anything that requires complex functionality on the back end, or that touches something payment-related, needs the review of experienced programmers in order to work.
