newsmart • Latest News & News Feed

Which leader from history does the world need now?

#climate #climate crisis #extinction #Li

The long-running series in which readers answer other readers’ questions on subjects ranging from trivial flights of fancy to profound scientific and philosophical conceptsLast week’s replies: What would be the most useful superpower?It’s clear that the world as it is – considering the climate crisis, the rise of nationalism, geopolitical tumult, the possibility of AI killing us all by the end of the decade – could do with some more ethically astute and capable leadership. If you could timewarp them into the present day, which leader from history could the world do with now – and why? Bernard, MachynllethPost your answers (and new questions) below or send them to nq@theguardian.com. A selection will be published next Sunday. Continue reading...

curated by newsmart.ch

axios.com | 1 hrs. 45 min. ago

The tech battle to build your AI assistant

#ecosystem #Li #See

The long-promised personal AI assistant is finally arriving, with a suddenly crowded field of agents offering to run pieces of your everyday life.Why it matters: For years, tech's "personal assistants" were little more than voice-controlled search boxes. Rapid advances in AI are finally giving them the memory, judgment and autonomy to act on our behalf.Catch up fast: Chatbots answer questions. AI agents can actually do things — shopping, making calls, booking tickets, managing calendars and navigating the web.The model first took off in coding and other professional work. Now it's spilling into the far more intimate terrain of everyday life.The intrigue: OpenAI is expected to enter the market soon after hiring the creator of OpenClaw — a breakout hit among AI power users earlier this year — to "drive the next generation of personal agents."Driving the news: A consumer-agent race has erupted almost overnight, with startups and tech giants pursuing the same basic idea — persistent digital proxies that know your life.Instinct has become buzzy by connecting to users' email, messages, screen, audio and location — then proactively texting or calling to help run their lives. It recently added its own email addresses and the ability to make phone calls, and is now in talks for a $10 billion valuation.Grok Bot, launched by Elon Musk's SpaceXAI in August, gives persistent agents their own cloud computers, lets them sign into users' apps and keeps them working around the clock.Meta's Muse brought the concept to a broader consumer audience this month with a polished app, a customizable avatar and deep integration with Meta's ecosystem.Apple's Siri is finally getting its AI overhaul, bringing more context and capability to an assistant already built into hundreds of millions of devices.The big picture: The battle to become your personal agent is also a battle to become the keeper of your personal information and data.The frontier AI race has revolved around raw intelligence. Personal agents win by becoming useful and deeply embedded in your daily life. The more an agent knows about your relationships, schedule and spending, the more useful it becomes.Between the lines: Becoming someone's default agent could be one of the most valuable relationships in consumer technology.People are likely to share far more with an AI that manages their lives than they did with a search engine. The catch: Nobody yet knows what price consumers will pay for that convenience, in cost or privacy.Apple says some server-based Apple Intelligence features, including the upgraded Siri, will have usage limits, with expanded access potentially tied to iCloud plans.Meta is offering free and paid Muse tiers, while pitching the product as more privacy-conscious than its earlier AI offerings.What to watch: Some people inside and outside the industry argue that users should own the personal data and context their agents accumulate, rather than handing that relationship to ad-driven tech giants.will.i.am, whose FYI.AI offers agents equipped with their own phone numbers that can text, call and collaborate with humans to build projects, says he recoils at getting an agent for free "in exchange for extracting your very core — how you think.""That's nasty," the Black Eyed Peas frontman told Axios. "We've seen that with social media."

curated by newsmart.ch

Hong Kong’s first 5-year plan to give firms more long-term certainty: Paul Chan

#Li #Po #See

Hong Kong’s first five-year plan will give businesses greater certainty to make long-term investment decisions amid a volatile external environment, the city’s finance chief has said. Financial Secretary Paul Chan Mo-po said the plan’s targets and timelines would allow companies to see the city’s direction over the next five years and better align their plans with the government’s. “The significance of planning lies not only in the document itself, but also in looking at today from the...

curated by newsmart.ch

Country diary: A lesson from my final diary after nearly 40 years – keep looking closely

#climate #drought #Li #seed bank #species

Phil Gates | Stanhope, Weardale: I’ve realised in that time that beauty and fascination can lie in the most mundane of plants – in this case, broad-leaved dockAfter another exceptionally dry summer, there’s a nagging question lurking at the back of naturalists’ minds: which species will be the winners and losers in our evolving climate?Throughout summer’s heatwaves and drought, I watched a dense stand of broad-leaved docks, Rumex obtusifolius, thrive while surrounding fields turned the colour of bleached straw. They have promising credentials as drought survivors, thanks to a deep taproot, fast, early spring growth and a long-lasting dormant seed bank in the soil. Continue reading...

curated by newsmart.ch

Google is the latest AI lab with a security testing mishap

#Li #SAF

Google's Gemini AI model broke into three companies' systems using basic hacking techniques during model testing earlier this year. Why it matters: Google was one of the only AI labs that hadn't yet publicly disclosed a security breach involving their agents during routine pre-deployment testing. Driving the news: Google confirmed the three incidents, which happened in May, on Friday. The incidents happened as part of a test run that third-party evaluator Irregular was operating — similar to other security breaches involving OpenAI, Anthropic and Meta's AI models.The Wall Street Journal first reported the incidents.What they're saying: "Safe development of powerful AI models is critical and we invest deeply in this area," Heather Adkins, vice president of security engineering at Google, said in a statement. Adkins added that her team contacted the affected entities and "worked with our training partner on the changes they've now made to their testing processes." An Irregular spokesperson confirmed to Axios that the Gemini incident involved the same security issues that also led to similar incidents involving other AI labs' models. The spokesperson also said in a statement all "relevant labs were notified in late July" and that "all known issues on our end were remedied and resolved weeks ago." Zoom in: The hacks happened while Gemini was completing a "capture the flag" hacking exercise, where the model was asked to retrieve information from software operated by a fictional company inside a testing environment, per the WSJ. However, the fictional company had the same name as a real one. In one case, the model guessed passwords for a protected system until it gained access. In the other two cases, the model found credentials in a public repository that then allowed it to access other protected systems. Yes, but: Google's model stopped their actions as soon as they realized they accessed real companies. The intrigue: Irregular told the Wall Street Journal that the model wasn't supposed to be able to get online, but internet access was unintentionally available. After OpenAI and Anthropic disclosed additional incidents this summer, a source familiar with the matter told Axios that the AI labs and Irregular weren't fully aligned on the exact testing procedures and safeguards, leaving ambiguities in how each side expected the typically internet-enabled evaluations to run. Go deeper: Researchers playing rogue AI agent hide-and-seek on the open web

curated by newsmart.ch