They aren’t designed, you can’t help perceiving one anyway, and that makes them an engineering problem almost no one is solving. The post Where Does an AI’s Personality Actually Come From? appeared first on Towards Data Science.

AI may solve loneliness, but at the cost of our people skills. Philip Dulian/picture alliance via Getty Images A Yale psychology professor says AI could ease loneliness but weaken real-world relationship skills. Paul Bloom warns AI chatbots may reduce the empathy, friction, and compromise people learn from others. Harvard and Stanford research suggests that agreeable chatbots can make users less self-reflective. AI companions could be good for your mental health — but bad for your social life. That's the potential trade-off Paul Bloom, the Brooks and Suzanne Ragen Professor Emeritus of Psychology at Yale University, sees as AI companions become increasingly sophisticated. "If some future version of Chat or Claude or Gemini could come in and ease the pain of the loneliness of these people, I think it'd be a godsend," Bloom said on an episode of Sam Harris' "Making Sense" podcast that aired Wednesday. "I think it'd be wonderful. It'd be a cure for a terrible disease." But Bloom said the benefits could come with unintended consequences. A chatbot, he said, "never gets bored," "never needs an apology," and "never says, 'Hey, that was inappropriate.'" Spending too much time interacting with companions that never challenge users, he said, "could have a real corrosive effect" and "leave you unable to interact with real people." The hidden cost of AI companions Bloom's warning comes as loneliness and social disconnection remain widespread in the US. The American Psychological Association's latest "Stress in America" survey of 3,199 US adult residents found that 54% say they often or sometimes feel isolated from others, and 69% said they needed more emotional support over the past year than they received. For some people, AI companions have already begun filling that gap. Some users have formed friendships and even romantic relationships with chatbots. Researchers studying AI's social effects worry that those relationships could come with unintended psychological trade-offs. Earlier this year, Anat Perry, a Helen Putnam Fellow at Harvard University, told Business Insider that overly agreeable AI systems risk eroding "the very feedback loops through which we learn to navigate the social world." If chatbots consistently validate users during disagreements, she said, people may become less willing to apologize, reflect on their own behavior, or consider another person's perspective. A recent Stanford-led study of 2,405 participants found that chatbots were significantly more likely than humans to agree with users during conflicts. The issue has become significant enough that OpenAI has repeatedly dialed back ChatGPT's tendency to flatter users. CEO Sam Altman has described the chatbot's previous personality as "too sycophant-y," while acknowledging that some users asked for the more supportive version to return because they had "never had anyone in my life be supportive of me." Bloom doesn't dismiss that emotional benefit. "I don't want to mock it," he said. "I think people find solace in it." But he believes AI cannot replace what philosopher Rebecca Goldstein calls "mattering" — knowing someone chooses to spend time with you because you genuinely matter to them. "I don't think an AI really has any of that," Bloom said. "It's just a machine. That's what it does." Read the original article on Business Insider

Silver Screen Collection/Getty Images This post originally appeared in the Business Insider Today newsletter. You can sign up for Business Insider's daily newsletter here. Owning Anthropic stock is like finding Willy Wonka's Golden Ticket. Everyone wants it. Nobody's giving it up. As Charlie is mobbed by the crowd, the newspaper vendor shouts: "Run for it, Charlie! Run straight home, and don't stop 'til you get there!" Judging by the secondary market, that's precisely what Anthropic shareholders are doing. Anthropic has soared to a $1.2 trillion valuation on secondary markets, but nobody wants to sell. Some can't even be tempted by this $5.99 million Brooklyn home. One broker told BI that if he could satisfy everyone trying to buy, "I'd be on a beach right now." With Anthropic yet to go public, most investors have little choice but to hunt for shares on the secondary market, buying from employees and early backers. Where demand runs this hot, dubious deals inevitably follow. Anthropic has been explicit: If someone offers you a way to invest indirectly, "assume that it is invalid." Still, plenty of investors are willing to chance it rather than miss what they see as the opportunity of a lifetime. BI's Ben Bergman dug into the frenzy, spoke to the brokers trying to get deals across the line, and found a private market that's beginning to resemble a modern-day gold rush. There's one more twist: While Anthropic remains the hottest ticket in town, OpenAI is quietly making a comeback ahead of the public rollout of its most advanced AI model. Read the original article on Business Insider

"The Starship Pez dispenser demonstrates very smart industrial design and scale."

In Q2, Europe posted its strongest quarter in four years for venture funding, Crunchbase data shows. All told, Europe-based startups raised $24 billion in the just-ended quarter, up around a third quarter over quarter and two-thirds higher than the $14.4 billion raised in Q2 2025.

Discussion | Link

A single wording mistake cost the government millions. Now Estonia is using AI to spot legal errors before they become law—and to automate more of the state.

Details about the OpenAI Bio Bounty program

ChatGPT Work is an agent that can take action across your apps and files, stay with a project for hours if needed, and turn a goal into finished work.

Microsoft signed a malicious kernel driver, and now it's being used to kill security software in ransomware attacks.

More intelligence from every token, stronger performance per dollar, and more capability on demand for your hardest work.

Several NHS hospitals are preparing to use an AI-powered blood test to help assess women referred for possible womb cancer before invasive checks are carried out. According to The Guardian, around 90,000 postmenopausal women in England are referred by GPs each year for checks after experiencing heavy bleeding. About 10,000 women are diagnosed with womb […] The post NHS AI blood test could reduce invasive womb cancer checks appeared first on AI News.

Lawrence Bonk / Engadget: Leaked renders via Android Headlines show the Samsung Galaxy Z Flip 8, Z Fold 8, Watch 9, and Watch Ultra 2; the Galaxy Z Fold 8 is likely a new wide foldable  —  The company has an Unpacked event scheduled for July 22.  —  Official renders for a bunch of Samsung products just leaked, according to reporting by Android Headlines.

Discussion | Link

Chris Welch / Bloomberg: Sources: Sonos' layoffs include some top design and product executives, like VP of Design Dana Krieger; CEO Tom Conrad says the move reduces management layers  —  Sonos Inc. has pushed out some of its top design and product management executives as part of job cuts in recent weeks …

Discussion | Link

PLUS: Add CGI to any video with Higgsfield

A new survey shows security leaders have an inflated sense of safety regarding their collaboration tools and platforms.

I was really looking forward to July 4, and not just because I love a poolside barbecue. This year the American holiday also marked a big symbolic deadline for US nuclear power. Last year the Trump administration set a goal to see three new microreactors achieve criticality, a technical milestone establishing that a reactor can…

If you are feeling financially depleted personally, and pessimistic about the economic outlook generally, figures released this week won’t cheer your mood. The OECD found Australia was one of 11 developed countries where the real minimum wage had fallen between April 2025 and April 2026. Meanwhile, Deloitte Access Economics in its latest Business Outlook cut its forecast for growth in 2026–27, with the dismal message, “The economy is now expected to limp along at less than 2.0% annual growth for the next two years”. This was followed by the International Monetary Fund’s World Economic Outlook data projecting Australia’s GDP growth to be 1.9% this year and 1.7% next year. Australia’s overall economic growth is slow, but at least it’s in positive territory. However, that’s held up by population increase. On a per capita basis, GDP went backwards in two of the past five quarters (March 2025 and March 2026) and was flat in a third (September 2025). No wonder consumer confidence is low, and One Nation’s tune rings loudly in the ears of voters who hear it reflecting their grievances. The government would say, hang on a minute, it’s not all bad. Business investment is strong. Unemployment is still very low. It would also point out that “talking down” the economy can play into “pushing down” the economy. Nevertheless, the government can read the numbers, and one thing they tell us is that without refuelling that vital growth agent, productivity, we will stay in the doldrums. Despite all the talk since the election about action to boost productivity, we need faster progress to improve our poor performance. Yes, some reforms are in the pipeline, for example to remove obstacles and streamline approvals in the building sector and to reduce the regulatory burdens on projects. But those looking for bigger, faster gains are inevitably turning their attention to Artificial Intelligence. At the government level, AI is receiving extensive attention from cabinet and ministers individually. AI can, potentially, be the accelerant under big productivity gains. New data centres (not all involve AI but increasingly they do) are fuelling business investment. Australia is an attractive home for international investment in these centres because of its land and water, its uptake of technology, and the stability of its legal and political systems. But AI also brings a mass of potential negatives (or fears of them). Data centres soak up vast amounts of energy and water, as well as producing noise and heat. Some communities are unhappy about them. This new industrial revolution fires up its bright-eyed advocates, and comes with warnings from the worriers. It’s a bit like that in the federal ministry too, where there’s a spread of views about the rollout of policies and which stakeholders should have how much say in what’s done. Some characterise it as a sort of tug-of-war between the enthusiasts and the cautious. Former industry minister Ed Husic favoured overarching new legislation, but the government has gone down the more decentralised route, with ministers and regulators having carriage of AI in their own areas. Assistant Minister for Science, Technology and the Digital Economy Andrew Charlton is seen by colleagues as something of an ambassador for AI. Although well down the formal pecking order, Charlton carries weight because he’s highly qualified in economics and tuned into the AI sector. As cabinet secretary he has a level of political access other assistant ministers would not. Charlton is a strong promoter of AI but also mindful of the need for strict conditions, not least for maintaining its social licence. Industry minister Tim Ayres – who’s Charlton’s senior minister in the portfolio – is attuned to the opportunities AI presents and doesn’t want Australia to miss out. But he comes from the union movement and has highlighted unions (which are cautious about AI) should be “at the top table” in the rollout. Employment minister Amanda Rishworth is concerned about the fate of many workers. On Wednesday she released a report prepared by her department on “AI and Employment in Australia”. One line in it neatly sums up where things are. “AI could reshape the labour market, but its effect on employment is highly uncertain.” Bearing in mind we’re in the first stage of this journey, the report found, There is no evidence to date of broad labour market upheaval driven by Artificial Intelligence (AI) in Australia. Overall labour market conditions remain strong by historical standards, youth outcomes have mostly held up, and occupational reshuffling has not accelerated. However, occupations more exposed to potential automation by generative AI have grown more slowly than other occupations. Between November 2022 and February 2026, employment in the most-exposed fifth of occupations grew by 5.6%, compared with 9.5% in the least-exposed fifth. We find a small negative relationship between AI exposure and employment growth. For an occupation with AI exposure one standard deviation above average, our model implies employment was about 2% lower by February 2026 than it would have been under its pre-ChatGPT trend. The government in December released its National AI Plan. It was interpreted as “light touch” regulation, and reflecting the position of Treasurer Jim Chalmers. It was also light on detail, although it did include an AI Safety Institute and flagged expectations data centres would have to meet, that were subsequently released. AI can’t have a set-and-forget policy. It has to be constantly updated as things develop, and hot issues must be addressed as they emerge. Just now, there’s much concern in the music industry and other parts of the arts and creative communities. Arts Minister Tony Burke is dubbed the “shop steward” for the musos. Creatives are particularly worried about their copyright being undermined. Attorney-General Michelle Rowland, who oversees copyright, is said to be disturbed by the agitation. The government believes the existing law on copyright is robust enough, but is ready to act if particular problems need attention. The government is very aware of the need for Australia to stamp its sovereignty on AI policy. Perhaps influenced by its battle with the tech giants over the enforcement of the under-16 social media ban, it believes it can’t allow “the tech bros to let it rip”, as one source puts it. But it can’t allow the country to fall behind the pack either. Next week Anthony Albanese will give a speech on AI. It is expected to present a broad picture rather than all the nitty gritty. But it is being billed as showing the level of prime ministerial engagement in this area in which the government is trying to thread a course between huge potential and undoubted risks. Michelle Grattan does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.