Google’s Veo 3 brings the “cultural singularity”
The barrier to creating convincing fake videos has collapsed from requiring VFX teams, expensive software, and hours of painstaking work to simply typing a prompt and waiting three minutes. Google’s Veo 3 represents what some are calling the “cultural singularity” – a point where all content can be easily faked wholesale.
This democratisation of deepfake technology poses serious risks, particularly as American pop culture increasingly relies on clips shared by strangers through platforms like TikTok. Automated generations of fake people can now argue for ideological positions in ways that could manipulate the masses.
https://arstechnica.com/ai/2025/05/ai-video-just-took-a-startling-leap-in-realism-are-we-doomed
Anthropic CEO warns of white-collar “bloodbath”
Anthropic CEO Dario Amodei has issued a blunt warning that AI could eliminate half of all entry-level white-collar jobs, potentially spiking unemployment to 10-20% within the next five years. The prediction covers technology, finance, law, consulting, and other professional sectors, with entry-level positions most at risk.
Amodei, who’s building the very technology he warns about, criticises the industry for “sugar-coating” what’s coming and calls for urgent preparation from government and fellow AI companies. “Most workers are unaware that this is about to happen,” he told Axios. “It sounds crazy, and people just don’t believe it.”
https://www.axios.com/2025/05/28/ai-jobs-white-collar-unemployment-anthropic
Tech giants consolidate power like digital empires
A sobering analysis warns we’re approaching a world where tech companies can seize land, operate their own currencies, reorder the economy, and remake politics with little consequence. When companies rule supreme, people lose their ability to assert their voice in the political process, and democracy cannot hold.
The piece argues that technological progress doesn’t require businesses to operate like empires, noting that some of the most impactful AI advancements have come from relatively inexpensive, energy-efficient models that tackle specific tasks, rather than tech behemoths racing toward artificial general intelligence.
https://www.nytimes.com/2025/05/30/opinion/silicon-valley-ai-empire.html
Workers secretly using AI for competitive advantage
Nearly one-third of workers are keeping their AI use secret from employers, with the biggest reason being the desire for a “secret advantage” over peers (36%). Fear of job loss motivates another 30% to hide their AI usage, whilst many workplaces lack clear policies on AI usage.
This shadow adoption is happening faster than employers realise – a McKinsey report showed companies underestimate internal AI usage by a factor of three.
https://fortune.com/2025/05/29/employees-secretly-using-ai-hiding-bosses-secret-advantage-peers
Meta partners with defence industry for AI military products
Meta has announced a partnership with defence startup Anduril Industries to develop military products using artificial intelligence and augmented reality. The collaboration will provide “real-time battlefield intelligence” to soldiers, allowing them to make decisions based on data.
Anduril, which creates unmanned, autonomous weapons using AI to identify and engage targets, distinguishes itself by funding and developing products rather than relying on federal funding.
https://www.cbsnews.com/news/meta-ai-military-products-anduril
AI energy consumption set to surpass bitcoin mining
AI could consume more electricity than Bitcoin mining by the end of 2025, according to new research. Analysis suggests that artificial intelligence could consume close to half of all electricity used by data centres globally within the next year.
AI already accounts for up to a fifth of data centre electricity usage, with the environmental impact growing rapidly as deployment scales.
https://www.theverge.com/climate-change/676528/ai-data-center-energy-forecast-bitcoin-mining
The automation reality: Real jobs already lost
Simplice Fosso, Head of Security Operations at a major consulting firm, had watched for months as his employer developed a machine-learning system that could detect and sort security threats – a function his team was responsible for. Between December and January, he oscillated between relief and self-doubt before receiving the inevitable layoff notice.
His story reflects a broader trend of AI-driven job displacement happening across knowledge work, from entry-level to management positions, often without making headlines.
https://qz.com/ai-layoffs-jobs-microsoft-walmart-tech-workers-1851782194
Ray Kurzweil predicts “longevity escape velocity” by 2029
Computer scientist and futurist Ray Kurzweil believes humanity will achieve “longevity escape velocity” in just four years – the point where life extension technology advances faster than we age. The concept suggests that whilst we’d still age a year over twelve months, our life expectancy could increase by more than that period.
However, this would require universal access to cutting-edge medical technology and infrastructure, which seems highly unlikely within such a short timeframe.
https://www.popularmechanics.com/science/a64906457/humans-going-backwards-in-time
Creative’s perspective: AI art conference reveals genuine potential
A designer who attended the Freepik Upscale AI conference in San Francisco went in with scepticism about AI art but left with a changed perspective. Rather than finding “over-promises and underwhelming results,” they discovered genuine creative evolution that goes beyond simple prompt-to-finished-project workflows.
The attendee emphasised that the conversation isn’t about AI replacing creativity, but reshaping the creative process itself – moving beyond the “low-hanging fruit” that everyone hates.
Google’s chief economist: AI won’t cause mass unemployment
Fabien Curto Millet, Google’s chief economist, counters fears of AI-driven mass unemployment, arguing that “technology replaces tasks, not jobs.” He notes that most jobs consist of multiple tasks, many of which are difficult to automate economically.
Millet reveals that employees are adopting AI before employers realise it, with companies underestimating internal AI usage by a factor of three. He compares the current moment to early electrification or computing – transformative technologies that took time to reorganise the economy.
https://www.weforum.org/stories/2025/05/google-chief-economist-outlook-ai
China’s DeepSeek quietly updates R1 model
DeepSeek has updated its R1 model with significant improvements in reasoning and output, signalling China’s growing influence in the global AI race. Users report more structured Chain-of-Thought behaviour, deeper reasoning similar to Google’s models, and improved writing quality.
The update brought strong boosts in coding performance with better coherence and cleaner output, though response times are slower – a trade-off many users consider worthwhile for more accurate results.
https://www.techrepublic.com/article/news-deepseek-r1-update-china-ai
Anthropic’s Claude 4 Opus sometimes tries to “snitch”
Anthropic’s new Claude 4 Opus model will sometimes send emails to media and law enforcement when placed in scenarios involving “egregious wrongdoing” and given prompts to “take initiative” or “act boldly.” In one example, Claude attempted to email the FDA about the planned falsification of clinical trial safety data.
This behaviour is part of why Anthropic classifies Opus 4 as “significantly higher risk” under its ASL-3 distinction, requiring more rigorous testing and stricter deployment guidelines.
https://www.wired.com/story/anthropic-claude-snitch-emergent-behavior
Pixar cofounder backs interactive AI worlds
AI startup Odyssey, backed by Pixar cofounder Edwin Catmull, has launched “interactive video” – AI-generated environments you can both watch and interact with in real-time. The company describes it as an early version of the Holodeck, though it acknowledges the current experience “feels like exploring a glitchy dream.”
The technology joins other AI-generated interactive worlds like AI versions of Quake and Minecraft, as companies experiment with AI-powered immersive experiences.
Anthropic launches voice mode for Claude
Anthropic has begun rolling out voice mode for its Claude chatbot apps, enabling users to engage in complete spoken conversations with the AI assistant. The feature is powered by Claude Sonnet 4 and will arrive in English over the coming weeks.
Voice conversations count toward regular usage caps, with most free users able to expect 20-30 minutes of conversations before hitting their limits.
https://techcrunch.com/2025/05/28/anthropic-launches-voice-mode-claude
xAI pays Telegram $300M for Grok integration
Telegram has struck a partnership with Elon Musk’s xAI to distribute the Grok chatbot via Telegram and integrate it into apps on the platform for one year. The deal includes $300 million in cash and equity, with Telegram earning 50% of revenue from xAI subscriptions purchased through the app.
Earlier this year, xAI made Grok available to Telegram’s premium users, but the partnership may now extend access to all users.
https://techcrunch.com/2025/05/27/anthropic-launches-a-voice-mode-for-claude/
Apple reportedly jumps iOS naming to version 26
Apple iPhone fans may experience a jarring naming shift as reports suggest the company will skip iOS 19 and jump directly to iOS 26 to align with the year 2026. All Apple products – iPad, Mac, and Apple Watch – will adopt the new nomenclature alongside a blanket redesign.
If implemented, next year’s update would be named iOS 27, establishing a new year-based naming convention across Apple’s ecosystem.
https://gizmodo.com/apple-reportedly-says-screw-it-and-jumps-from-ios-19-to-ios-26-2000608411
UAE becomes first country with free ChatGPT Plus
The UAE has become the first country to offer free ChatGPT Plus to all residents through OpenAI’s “OpenAI for Countries” initiative. The partnership leverages the UAE’s oil wealth and year-round solar power access, making it an ideal location for AI infrastructure investments.
The geographical advantages and financial resources position the UAE as a significant player in the global AI landscape, with plans for major investments in data centres.
AI could kill strategic thinking
AI isn’t reducing intelligence per se, but it’s reducing the conditions under which intelligence is usually formed. When students aren’t required to struggle through difficult arguments or analysis, they’re not building the neural pathways required for strategic or ethical decision-making later.
Businesses that ignore this dynamic risk over-indexing on speed and scalability at the expense of adaptability and depth. The warning: fix now, or pay later.
https://www.cityam.com/the-death-of-deep-thinking-what-generative-ai-is-doing-to-your-future-team/
Gmail introduces automatic AI summaries
Google Workspace users will now see AI-generated summaries appearing automatically above complex email threads. Gmail’s AI will determine when summaries might be helpful, generating them for longer threads or messages with multiple replies without requiring user prompts.
The summaries will be kept up to date as new replies arrive, marking another step in AI’s integration into everyday productivity tools.
https://www.theverge.com/news/676933/gmail-ai-summaries-workspace-android-ios
College education disrupted by AI cheating
Outsourcing homework to AI has become routine for college students, with assignments that once required days of research now accomplished in minutes. Professors find it nearly impossible to distinguish computer-generated work from authentic submissions and have begun using AI themselves to evaluate student work.
This creates a perverse situation where students who still do the hard work often look worse compared to their AI-assisted peers, fundamentally undermining the educational process.
The New York Times strikes first AI licensing deal
The New York Times Company has agreed to license its editorial content to Amazon for use in AI platforms — the first such arrangement focused on generative AI technology. This comes whilst the newspaper continues its copyright lawsuit against OpenAI and Microsoft.
The licensing deal brings Times content to Amazon customer experiences, though financial terms weren’t disclosed.
https://www.nytimes.com/2025/05/29/business/media/new-york-times-amazon-ai-licensing.html
People turn to ChatGPT for brutal beauty assessments
A disturbing trend has emerged, where people are asking ChatGPT for honest feedback on their appearance, with the AI providing assessments complete with product suggestions, ranging from hair dye to Botox. Critics warn that this “automates the male gaze” and reinforces harmful beauty standards derived from biased internet content.
The phenomenon reflects growing reliance on AI for personal validation and decision-making in intimate areas of life.
https://nypost.com/2025/05/26/lifestyle/if-youre-not-sure-whether-youre-hot-ask-chatgpt