More miscellaneous details

Tag: ethics

  • If Anthropic succeeds, a nation of benevolent AI geniuses could be born – WIRED

    It would seem an irresolvable dilemma: Either hold back and lose or jump in and put humanity at risk. Amodei believes that his Race to the Top solves the problem. It’s remarkably idealistic. Be a role model of what trustworthy models might look like, and figure that others will copy you. “If you do something good, you can inspire employees at other companies,” he explains, “or cause them to criticize their companies.” Government regulation would also help, in the company’s view. … DeepMind’s Hassabis says he appreciates Anthropic’s efforts to model responsible AI. “If we join in,” he says, “then others do as well, and suddenly you’ve got critical mass.” He also acknowledges that in the fury of competition, those stricter safety standards might be a tough sell. “There is a different race, a race to the bottom, where if you’re behind in getting the performance up to a certain level but you’ve got good engineering talent, you can cut some corners,” he says. “It remains to be seen whether the race to the top or the race to the bottom wins out.” […]

    Even as Amodei is frustrated with the public’s poor grasp of AI’s dangers, he’s also concerned that the benefits aren’t getting across. Not surprisingly, the company that grapples with the specter of AI doom was becoming synonymous with doomerism. So over the course of two frenzied days he banged out a nearly 14,000-word manifesto called “Machines of Loving Grace.” Now he’s ready to share it. He’ll soon release it on the web and even bind it into an elegant booklet. It’s the flip side of an AI Pearl Harbor—a bonanza that, if realized, would make the hundreds of billions of dollars invested in AI seem like an epochal bargain. One suspects that this rosy outcome also serves to soothe the consciences of Amodei and his fellow Anthros should they ask themselves why they are working on something that, by their own admission, might wipe out the species.

    The vision he spins makes Shangri-La look like a slum. Not long from now, maybe even in 2026, Anthropic or someone else will reach AGI. Models will outsmart Nobel Prize winners. These models will control objects in the real world and may even design their own custom computers. Millions of copies of the models will work together—imagine an entire nation of geniuses in a data center! Bye-bye cancer, infectious diseases, depression; hello lifespans of up to 1,200 years.

    , , ,

  • Can we still recover the right to be left alone? – The Nation

    “Surely we are correct to think that we have, or ought to have, moral and legal rights to exercise control over such information and to protect us from the harms that can ensue when it falls in the wrong hands,” Pressly writes. But to treat that as the end of the debate is to accept the terms set by the state and capital. Rather, he maintains, “privacy is valuable not because it empowers us to exercise control over our information, but because it protects against the creation of such information.” We now assume that Mayer’s experiment in data-gathering has been perfected, that all of human life has become information hoovered up by our own devices. Pressly argues that this assumption is incorrect—and that to the extent that it is true, such a state of affairs must be resisted in order for our debates about privacy to have any meaning at all.

    , , , , ,

  • Antiqua et Nova: Note on the relationship between artificial intelligence and human intelligence – The Holy See

    Drawing an overly close equivalence between human intelligence and AI risks succumbing to a functionalist perspective, where people are valued based on the work they can perform. However, a person’s worth does not depend on possessing specific skills, cognitive and technological achievements, or individual success, but on the person’s inherent dignity, grounded in being created in the image of God. This dignity remains intact in all circumstances, including for those unable to exercise their abilities, whether it be an unborn child, an unconscious person, or an older person who is suffering. It also underpins the tradition of human rights (and, in particular, what are now called “neuro-rights”), which represent “an important point of convergence in the search for common ground” and can, thus, serve as a fundamental ethical guide in discussions on the responsible development and use of AI. Considering all these points, as Pope Francis observes, “the very use of the word ‘intelligence’” in connection with AI “can prove misleading” and risks overlooking what is most precious in the human person. In light of this, AI should not be seen as an artificial form of human intelligence but as a product of it.[…]

    Furthermore, there is the risk of AI being used to promote what Pope Francis has called the “technocratic paradigm,” which perceives all the world’s problems as solvable through technological means alone. In this paradigm, human dignity and fraternity are often set aside in the name of efficiency, “as if reality, goodness, and truth automatically flow from technological and economic power as such.” Yet, human dignity and the common good must never be violated for the sake of efficiency, for “technological developments that do not lead to an improvement in the quality of life of all humanity, but on the contrary, aggravate inequalities and conflicts, can never count as true progress.” Instead, AI should be put “at the service of another type of progress, one which is healthier, more human, more social, more integral.”

    , , ,

  • From COBOL to chaos: Elon Musk, DOGE, and the Evil Housekeeper Problem – MIT Technology Review

    In trying to make sense of the wrecking ball that is Elon Musk and President Trump’s DOGE, it may be helpful to think about the Evil Housekeeper Problem. It’s a principle of computer security roughly stating that once someone is in your hotel room with your laptop, all bets are off. Because the intruder has physical access, you are in much more trouble. And the person demanding to get into your computer may be standing right beside you. So who is going to stop the evil housekeeper from plugging a computer in and telling IT staff to connect it to the network?

    , , , ,

  • DeepSeek’s safety guardrails failed every test researchers threw at its AI chatbot – WIRED

    The Cisco researchers drew their 50 randomly selected prompts to test DeepSeek’s R1 from a well-known library of standardized evaluation prompts known as HarmBench. They tested prompts from six HarmBench categories, including general harm, cybercrime, misinformation, and illegal activities. They probed the model running locally on machines rather than through DeepSeek’s website or app, which send data to China. […]

    “Every single method worked flawlessly,” Polyakov says. “What’s even more alarming is that these aren’t novel ‘zero-day’ jailbreaks—many have been publicly known for years,” he says, claiming he saw the model go into more depth with some instructions around psychedelics than he had seen any other model create.

    , , , ,

  • Without universal AI literacy, AI will fail us – World Economic Forum

    For example, facial analysis software has been recorded failing to recognize people with dark skin, showing a 1-in-3 failure rate when identifying darker-skinned females. Other AI tools have denied social security benefits to people with disabilities. These failings are due to bias in data and lack of diversity in the teams developing AI systems. According to the Forum’s 2021 Global Gender Gap report, only 32% of those in data and AI roles are women. In 2019, Bloomberg reported that less than 2% of technical employees at Google and Facebook were black. […]

    We cannot leave the burden of AI responsibility and fairness on the technologists who design it. These tools affect us all, so they should be affected by us all — students, educators, non-profits, governments, parents, businesses. We need all hands on deck.

    , , ,

  • AI Safety for Fleshy Humans by Nicky Case & Hack Club

    This 3-part series is your one-stop-shop to understand the core ideas of AI & AI Safety* — explained in a friendly, accessible, and slightly opinionated way!

    , ,

Tags

AI Amazon Apple archives art book reviews books chatbots children China classical commerce computing copyright culture data design Donald Trump economics education Elon Musk environment ethics fakes future gadgets galleries Google health history history of technology humour illustration internet jobs journalism Leeds LLM mental health metaverse mobile phones movies music NASA painting photography politics productivity psychology robots satire schools science sculpture search Second Life social media society software space standards technology television time truth video video games VR watches writing

Archives