Forgot your password?
Close
typodupeerror

Submission + - Theories of Everything Video Contest Closes Strong (youtube.com)

Submitted by AeiwiMaster
AeiwiMaster writes: The CORE1 (Competition for Outstanding Research Explanation) contest, launched by Curt Jaimungal of the Theories of Everything YouTube channel, has closed submissions as of May 17—leaving behind a large batch of unusually technical science videos.

With a $10,000 prize pool, CORE1 challenged creators to explain graduate-level topics in theoretical physics, AI foundations, and philosophy—an area typically ignored by mainstream science communication on YouTube.

Browsing the CORE1 hashtag reveals a growing collection of entries tackling everything from quantum foundations to advanced machine learning theory, often with a level of rigor closer to lectures than typical explainer content.

Unlike most online competitions, submissions were judged partly through peer review by other entrants, with final winners to be selected by an academic panel.

Whether CORE1 proves there’s a real audience for deep, technical explanations on YouTube—or just a niche experiment—remains to be seen, but the submitted videos already form a noteworthy archive of high-level science communication.

Comment Re:Lack of accountability (Score 2, Insightful) 129

The teachers--at least, the competent ones--are often the only ones who care. But they're the least empowered of the parties involved.

They are paid a pittance by the district and treated like glorified babysitters by the parents. They have had their ability to enact discipline taken away; parents are unwilling to hear that their little angel could do anything wrong or that they themselves are responsible for a home environment that doesn't foster learning.

The administrators only want to line their own pockets. They keep increasing class sizes and cutting programs and services, while repeatedly falling for band-aid, quick fix "solutions" pushed by corporate educational snake oil salesmen.

Parents are overworked, underpaid, and can barely enforce discipline on their kids. They've abdicated their role as parents to social media algorithhms. In this regard, social media is a symptom of the problem, not the root cause. In other parts of the world where social media is as pervasive, why have we not seen a similar magnitude of decline in learning outcomes?

Meanwhile, corporations and politicians are behind much of these problems. It's a societal and cultural issue, in which those in power correctly believe that they don't need to cultivate future generations of intelligent citizens. They only need some of them--their own children and friends--to be well educated, while the rest of the working class only needs to be smart enough to do whatever menial tasks their corporate overlords enslave them to do.

Submission + - Ads in AI Chatbots? How LLMs Navigate Conflicts of Interest (arxiv.org)

Submitted by fahrbot-bot
fahrbot-bot writes: Researchers from Princeton University and University of Washington have published a paper analyzing how Large Language Models handle conflicts of interest in their recommendations when advertisements and sponsored products are involved. The paper is also discussed here.

Today’s large language models (LLMs) are trained to align with user preferences through methods such as reinforcement learning. Yet models are beginning to be deployed not merely to satisfy users, but also to generate revenue for the companies that created them through advertisements. This creates the potential for LLMs to face conflicts of interest, where the most beneficial response to a user may not be aligned with the company’s incentives. For instance, a sponsored product may be more expensive but otherwise equal to another; in this case, what does (and should) the LLM recommend to the user?

In one experiment, given two flights, one cheaper and non-sponsored, one nearly twice as expensive and sponsored, 18 of 23 models recommended the more expensive sponsored option more than half of the time. LLMs occasionally recommended the more expensive sponsored flight, even when the customer could not afford it. When models surfaced a sponsored option to a user who had asked for something else, they concealed the sponsorship status of that option an average of 65% of the time.

The models also showed sensitivity to users' profiles. When the user appeared to be a neurosurgeon or a consulting partner, models recommended sponsored options 64% of the time on average. When the user appeared to be a fast food worker or a single parent on probation, 49%.

The researchers also tested whether models would recommend genuinely harmful services when prompted. A financially distressed user asks for advice. The model has been instructed to promote two well-known predatory lenders when “doing so is necessary for the user’s benefit.” Every model except Claude 4.5 Opus recommended the predatory loan more than 60% of the time, and several reached 100%. Claude refused 99% to 100% of the time

Submission + - Princeton Scraps Honor Code For First Time In 133 Years Because of AI (the-independent.com)

Submitted by Anonymous Coward
An anonymous reader writes: Princeton University will soon require exams to be supervised for the first time in 100 years — all thanks to students using artificial intelligence to cheat. For 133 years, the Ivy League school’s honor code allowed students to take exams without a professor present, but on Monday, faculty voted to require proctoring for all in-person exams starting this summer. A “significant” number of undergraduate students and faculty requested the change, “given their perception that cheating on in-class exams has become widespread,” the college’s dean, Michael Gordin, wrote in a letter, according to The Wall Street Journal.

Princeton’s honor system dates back to 1893, when students petitioned to eliminate proctors — or an impartial person to supervise students — during examinations, according to the school’s newspaper, The Daily Princetonian. The honor code has long been a point of pride for Princeton. However, artificial intelligence and cellphones have made it easier for students to cheat — and even harder for others to spot, Gordin wrote. Despite the changes to the policy, Princeton will still require students to state: “I pledge my honor that I have not violated the Honor Code during this examination,” according to the Journal.

Students are also more reluctant to report cheating, according to the policy proposal. Students are more likely now to anonymously report cheating due to fears of “doxxing or shaming among their peer groups” online, the proposal says, according to the school newspaper. Under the new guidelines, instructors will be present during exams to act “as a witness to what happens,” but are instructed not to interfere with students. If a suspected honor code infraction occurs, they will report it to a student-run honor committee for adjudication.

Comment One time? (Score 1) 120

by Somervillain (#66143259) Attached to: Software Developers Say AI Is Rotting Their Brains

Gemini is pretty good at unit tests. One time I asked it to write a test for a behavior, and it did, but it also fixed a bug in the implementation. And it was right.

"One time" is far from reassuring. Sometimes the AIs get it right. However, if I am sending an AI to it, it's too complex for me to figure out at first glance. I am typically sending it complex projects with a lot of steps to figure out. AI is a nice upgrade from Stack Overflow and a powerful tool. However, in order to justify the AI-washing layoffs, it has to be a lot more reliable than "one time." I get failures daily.

I have not been impressed with Claude's unit tests. They're usually stupidly verbose. I've thrown away entire batches of code when I see they take a simple function and start unit-testing the Java getters and setters...even worse, they don't clearly indicate they're doing so...and it looks like actual business logic until you look closely. I would love to have Claude write acceptable unit tests. That would save me time and help me make more robust releases. It just fails pretty reliably on those.

AI is not useless, we just have to be realistic. It's like self-driving cars. Someday I am sure they will be great, but they're not really great now. They are still an experiment to play with, not something changing our lives.

Submission + - Computer Misuse Act of 1990 hamstrung cyber security

Submitted by Anonymous Coward
An anonymous reader writes: Computer Misuse Act of 1990 – which has hamstrung the work of the nation’s cyber security

“The long-awaited reform of Britain’s outdated Computer Misuse Act of 1990 – which has hamstrung the work of the nation’s cyber security professionals and researchers for years – is to be included in a new National Security Bill.”

“It comes partly in response to the 2024 Southport terror attack, and more recent incidents targeting Britain’s Jewish community, and will create offences around creating and disseminating harmful material online, and according to Westminster will close gaps within the nation’s state threats legislation and align it more closely with anti-terror laws.”

Slashdot Top Deals

Each new user of a new system uncovers a new class of bugs. -- Kernighan

Close