Group items tagged

"Despite these promises, concerns over privacy, security, and exclusion are growing. Critics, including Privacy International and Access Now, have raised alarm over the risks posed by poorly regulated digital ID systems. They caution, "Without strong safeguards, Africa's digital ID dream risks becoming a surveillance nightmare." Data breaches and inadequate legal frameworks further raise questions about whether these programs empower citizens or expose them to new vulnerabilities."

"This is a broader pattern we're seeing across other fields where LLMs are being deployed. Whether it's coding, writing, design, law, or medicine, the most effective AI users are people who already have deep domain expertise. Expertise isn't obsolete; it's more important than ever-because the value isn't just in producing outputs quickly. It's in being able to vet, steer, and improve those outputs. The future of computer science education isn't about teaching less. It's about teaching differently. We still need students who can understand how software works at a fundamental level. But we also need to train them to collaborate with AI-to become fluent in prompting, reviewing, debugging, and refining AI-generated outputs. Mastering this hybrid skillset will be critical not just for engineers, but for anyone hoping to thrive in a world where knowledge work is increasingly AI-augmented. Practically speaking, AI could dramatically lower the barrier to entry for students. When I was in high school, it would take months (if not years) of training in CS before you could create a game or app that was genuinely cool to people that aren't inherently curious and nerdy."

"AI models will inevitably have to make value judgments. If we want those judgments to be congruent with our own values (which is, after all, the central goal of AI alignment research) then we need to have ways of testing which values a model expresses in the real world. Our method provides a new, data-focused method of doing this, and of seeing where we might've succeeded-or indeed failed-at aligning our models' behavior."

"Generative AI chatbots fail to adequately reflect fossil fuel companies' complicity in the climate crisis, a Global Witness investigation has found"

60% of TikTok users and 46% of Instagram users say they feel worse off because of these platforms. 57% and 58% of college students (users and non-users) prefer to live in a world without TikTok and Instagram, respectively. Users would even pay money to see them disappear-$24 on average to eliminate TikTok and $6 to eliminate Instagram.

"A rise in online racism driven by fake images is "just the start of a coming problem" after the latest release of X's AI software, online abuse experts have warned. Concerns were raised after computer-generated images created using Grok, X's generative artificial intelligence chatbot, flooded the social media site in December last year. Signify, an organisation that works with prominent groups and clubs in sports to track and report online hate, said it has seen an increase in reports of abuse since Grok's latest update, and believes the introduction of photorealistic AI will make it far more prevalent."

"Among kids under age 13 (i.e., 11- and 12-year-olds), they found: 63.8% reported using social media They had an average of 3.38 social media accounts Among the 63.8% with a social media account, TikTok was most popular, with 68.2% saying they have an account. 57.3% said they have Instagram and 55.2% Snapchat Only 5.4% said their social media account(s) were secret from their parents My take: The takeaway here is simple: a lot of kids are using social media! This data may slightly overestimate the numbers, given data collection during the pandemic (2019-2021), when rates of social media use may have been higher. But still: 64% using social media! Despite a national desire to put our fingers in our ears and scream "la-la-la," kids under 13 are using these platforms. We need to either do a better job of preventing that, or make the platforms safer for kids that age. Academic Pediatrics."

"Instead, we need to preserve and protect these personal interactions. We need to bolster the working conditions of connective labour practitioners so they are able to see others well. We need to impose a "connection criterion" to help us decide which AI to encourage - the kind that creates new antibiotics, for instance, or decodes sperm whale language - and which to put the brakes on, that is, the kind that intervenes in human relationships. Each of us needs to decide how much we value the human connections in our lives and the lives of our neighbours."

"When employees at meta started developing their flagship AI model, Llama 3, they faced a simple ethical question. The program would need to be trained on a huge amount of high-quality writing to be competitive with products such as ChatGPT, and acquiring all of that text legally could take time. Should they just pirate it instead?"

"Our estimate of the length of tasks that an agent can complete depends on methodological choices like the tasks used and the humans whose performance is measured. However, we're fairly confident that the overall trend is roughly correct, at around 1-4 doublings per year. If the measured trend from the past 6 years continues for 2-4 more years, generalist autonomous agents will be capable of performing a wide range of week-long tasks."

"But what about social media posts that offer stories of hope and recovery? Could these types of posts actually prevent suicide? For this experimental study, researchers in Austria created 10 suicide-prevention social media posts from a fictitious influencer. The posts offered stories about recovery from suicidal crises, mental health tips, and life-affirming messages. A total of 354 adult participants were randomly assigned to view these posts, or to view 10 posts totally unrelated to mental health. As expected, participants who were exposed to the suicide-prevention posts reported decreased suicidal thoughts and greater intentions to seek help (e.g., from friends, family, or a professional). This was especially true for those who were already struggling with suicidal thoughts."

"I Used to Teach Students. Now I Catch ChatGPT Cheats I once believed university was a shared intellectual pursuit. That faith has been obliterated"

"At least 50 separate deadly strikes by armed forces in Africa have been confirmed during the three years up to November 2024, with analysts describing a "striking pattern of civilian harm" with little or no accountability. Although the rapid growth of armed drones deployed by Ukraine and Russia receives significant scrutiny, scant focus is being paid to the escalating use in Africa of a new breed of imported cheaper drones, such as Turkey's Bayraktar TB2, said Cora Morris of campaign group Drone Wars UK, which on Monday published a report on the growth of armed drones in Africa, called Death on Delivery."

"Umi is a digital wellness platform and personalized family wellness coach that integrates AI and a community of experts. By consolidating wellness data into personalized, actionable pathways for every family, Umi helps people build healthy habits and connect through wellness routines that work for everyone. And by leveraging the power of Anthropic's Claude AI model, Umi benefits from the advanced reasoning capabilities to navigate the unique challenges faced by individuals and families."

"Not everyone is excited for the change. Publishers are completely freaked out. The shift has heightened fears of a "zero-click" future, where search referral traffic-a mainstay of the web since before Google existed-vanishes from the scene.  I got a vision of that future last June, when I got a push alert from the Perplexity app on my phone. Perplexity is a startup trying to reinvent web search. But in addition to delivering deep answers to queries, it will create entire articles about the news of the day, cobbled together by AI from different sources. "

"Put more simply? We're in the midst of a digital slopaggedon, with the felt everyday impacts ranging from mild irritants like graphs about "uncoked kermelss" to more serious information erosions - the replacing of images of famous artworks and historical events with AI-generated fakes, for example, or the flooding of social media with fake AI-generated imagery during environmental disasters."

"This cycle is particularly painful for non-engineers because they lack the mental models to understand what's actually going wrong. When an experienced developer encounters a bug, they can reason about potential causes and solutions based on years of pattern recognition. Without this background, you're essentially playing whack-a-mole with code you don't fully understand."

"The announcement that Meta would be changing their approach to political content and discussions of gender is concerning, though it is unclear exactly what those changes are. Given that many product changes regarding those content areas were used in high-risk settings, a change intended to allay US free speech concerns could lead to violence incitement elsewhere. For example, per this post from Meta, reducing "content that has been shared by a chain of two or more people" was a content-neutral product change done to protect people in Ethiopia, where algorithms have been implicated in the spread of ethnic violence. A similar change - removing optimizations for reshared content - was discussed in this post concerning reductions in political content. Will those changes be undone? Globally? Such changes could also lead to increased amplification of attention getting discussions of gender. Per this report from Equimundo and Futures Without Violence, 40% of young men trust at least one "manosphere" influencer - who often exploit algorithmic incentives by posting increasingly extreme, attention-getting mixes of ideas about self-improvement, aggression, and traditional gender roles."

"What we know about reducing misinformation Where the evidence stands for six strategies used to reduce the spread and impact of misinformation, according to the Prosocial Design Network"