Hashtag Trending

A $20 YouTube promotion becomes a reported $118,000 company credit card bill. Canada's Cohere tackles a gap in AI data security. And could realistic scenarios help us understand—and prevent—an AI-enabled catastrophe? Jim Love hosts Hashtag Trending for Friday, September 18, 2026. A father says his nine-year-old ran up $118,000 advertising Minecraft and Roblox videos over roughly three weeks. We examine how a shared Google account and saved company card enabled the spending—and what that could mean for AI agents given purchasing authority. Cohere adds confidential computing to Model Vault to protect sensitive enterprise information while AI processes it. We explain why a promise not to train on your data doesn't address exposure in working memory, and how the technology is designed to block access even by Cohere and its cloud providers. Plus, Cohere signs a merger agreement with Germany's Aleph Alpha, subject to regulatory approval. Then, "Rumours of Our Death": CNN explores four AI extinction scenarios involving engineered pandemics, autonomous robots, nuclear weapons and control of infrastructure. What could AI contribute? What physical barriers remain? And what does our response to climate change suggest about our ability to agree on effective safeguards? CHAPTERS 00:00 Headlines and introduction 00:40 Father reports $118K in YouTube ad charges 01:31 Spending controls and AI agents 02:53 Cohere's Model Vault privacy protection 04:22 How confidential computing works 05:01 Cohere–Aleph Alpha merger agreement 05:29 Could AI wipe out humanity? 06:16 Four AI extinction scenarios 09:18 Current AI capabilities and misuse 10:24 Would we act to prevent catastrophe? 11:53 Could AI help us assess the risks? 12:16 Wrap-up and subscribe Subscribe for daily technology news, and join us Saturday for Project Synapse, where we take a deeper look at the week's AI developments. What safeguards would you require before giving an AI agent access to company data or spending authority? Leave a comment below. Contact Jim through the Contact Us page at technewsday.ca or technewsday.com.

On the 2,500th episode of Hashtag Trending (Thursday, September 17), host Jim Love welcomes former hosts and tech journalists Brian Jackson and Alex Coop. Brian previews Tech Trends 2027, warning CIOs about rising insider threats enabled by AI wearables and prediction markets, faster AI-assisted vulnerability exploitation that makes weekly/monthly patching obsolete, and geopolitical uncertainty affecting procurement of frontier AI services. Alex highlights SAS's second annual Data and AI Impact Report, showing organizations with strong governance, data quality, and auditability achieve at least double the AI ROI, while only 19.3% of Canadian respondents expect strong returns despite a rising AI trustworthiness index and growing compliance concerns. Jim also discusses research suggesting Americans spend about eight hours online daily and offers tactics to reduce screen-time triggers, followed by an afterword conversation reflecting on the show's origins, production challenges, milestones, and what the hosts do now. 00:00 2500th Episode Kickoff 01:18 Tech Trends 2027 Preview 02:11 Insider Threats and Wearables 03:31 AI Vulnerability Management 04:44 Geopolitics and Procurement 05:35 SAS Trust and ROI Report 07:51 Online Time and Attention Economy 11:24 Reclaiming Screen Time 12:15 Social Media Reflections 13:40 Afterword Trending Origins 19:51 Favorite Memories and Mishaps 24:01 Where They Are Now 29:15 Final Sign Offs

Who pays if the AI investment boom falls short—and who benefits if development slows down? Jim Love examines the money, motives and economic assumptions in this special AI Panic edition of Hashtag Trending. MIT Technology Review explores the risks behind roughly $750 billion in expected spending by the biggest cloud companies. We examine debt financing, competition from cheaper models such as DeepSeek, recurring chip replacement costs, and potential exposure for pension funds, life insurance investments and electricity customers. Business Insider questions the conveniently timed calls for an AI slowdown. Could safety regulations also protect established companies from competition? We discuss OpenAI ruling out a 2026 IPO, Anthropic's continuing listing plans, and the publicity surrounding researcher Jacob Coxon's resignation. Finally, we examine Anthropic's economic scenarios—including an extreme case in which US GDP reaches roughly 32% above its no-AI baseline by 2030. That is a scenario, not a prediction, and it comes with substantial unemployment, difficult retraining questions and significant gaps in what the model covers. CHAPTERS 00:00 AI Panic Edition: Headlines 00:42 The $750B AI Data Centre Gamble 03:32 AI Debt Risks: Dot-Com Bust or Another 2008? 05:24 Who Benefits From an AI Slowdown? 07:24 Jacob Coxon's Resignation and Safety Lobbying 09:23 Anthropic's 2030 GDP and Jobs Scenarios 11:46 Wrap-Up and How to Reach Us Can AI deliver real economic gains while its biggest investors lose money? And can safety concerns be sincere while also serving corporate interests? Share your thoughts in the comments. Subscribe to Hashtag Trending for technology news with Jim Love. Contact us through technewsday.ca or technewsday.com.

AI safety concerns are growing, but agreement on what to do remains elusive. Can governments establish effective oversight amid U.S.–China rivalry, industry self-interest and questions about transparency? On Hashtag Trending for Tuesday, September 15, 2026, Jim Love examines the push to slow AI development—and why both industry proposals and rushed legislation deserve scrutiny. Also in this episode: • Why companies postpone database upgrades. A Percona-sponsored survey produces a startling 97% headline, but the sample and meaning of "outdated" matter. We explore compatibility failures, testing costs and the difference between older software and software without security patches. • Frank Shaw's Microsoft departure. The longtime communications chief will leave at the end of December after helping shape Microsoft's message across the Bill Gates, Steve Ballmer and Satya Nadella eras. • A Canadian teenager puts AI to work for others. Toronto Grade 11 student Nathan Yizhou Han is developing an app to help people experiencing homelessness navigate housing, shelters and support services. His award-winning robotics experience offers a hopeful perspective on purposeful AI education. CHAPTERS 00:00 Today's Tech Headlines 00:38 AI Safety: The U.S.–China Standoff 03:05 Who Can We Trust on AI Safety? 03:59 AI Regulation: Making the Rules Work 05:03 Why Companies Run Outdated Databases 05:40 Database Upgrades: Costs and Compatibility 07:58 When Older Software Becomes a Security Risk 09:14 Frank Shaw Is Leaving Microsoft 10:50 Canadian Teen Builds AI Tool for Homeless Support 12:05 Robotics Awards and Canada's CanCode Programme 14:08 AI in Schools: Creativity and Compassion 15:20 Closing and Subscribe What would meaningful AI oversight look like to you? Leave a comment below. Subscribe to Hashtag Trending for technology news with context and a critical lens. Contact Jim through the Contact Us page: https://technewsday.ca https://technewsday.com

Sam Altman and Dario Amodei back slowing AI development as researchers raise fresh safety concerns. Microsoft brings Grok to Copilot, and a Flock camera investigation triggers more police arrests and firings. Jim Love covers these stories on Hashtag Trending for Monday, September 14, 2026: • AI safety and regulation: Former Anthropic researcher Joe Benton and former Google DeepMind researcher Josh Engels speak out following Jacob Coxon's resignation. OpenAI rules out a 2026 IPO, California advances independent AI oversight, and Bernie Sanders and Greg Casar propose restrictions on superintelligence. • Microsoft adds Grok: Copilot expands model choice in Word, Excel and PowerPoint. We examine Grok's pricing, workplace agents and Microsoft's growing range of alternatives to OpenAI. • Flock camera misuse: USA TODAY's reporting prompts arrests, firings and investigations. Communities suspend camera use or change suppliers—but who monitors how police use surveillance technology? • AI in Yemen: Houthi opponents allege a cloned voice circulated a false retreat order. Its authenticity and battlefield impact remain unverified. Separately, Anthropic reports that a weapons-development cell in northern Yemen used Claude for missile-guidance software and test analysis before its accounts were blocked. CHAPTERS 00:00 Today's tech headlines 00:57 AI researchers warn about loss of control 02:26 Altman, Amodei and the AI regulation push 04:53 Can the world agree to slow AI? 05:27 Microsoft adds Grok to Copilot 05:42 Grok pricing, Tesla and Starlink 07:17 Grok's AI teammates and Microsoft's strategy 08:14 Flock camera investigation triggers arrests 10:30 Who holds police surveillance accountable? 11:21 Yemen's alleged AI-generated retreat order 11:53 Anthropic uncovers Claude-assisted weapons work 13:19 Closing and sign-off Subscribe to Hashtag Trending for technology news with Jim Love. Share your thoughts in the comments: what would meaningful oversight of AI and surveillance technology look like? Contact us through the Contact Us page at technewsday.ca or technewsday.com.

OpenAI is cutting off Cursor's direct model access ahead of Astra, while IBM's claimed quantum advantage already faces a classical challenge. Jim Love covers the technology headlines for Monday, August 31, 2026. OpenAI will stop supplying models directly to Cursor on November 12 following Cursor's $60-billion acquisition by SpaceX. OpenAI cites its contract and concerns that model outputs could be used to help train competing xAI systems. Developers can still access OpenAI through standard APIs, Codex and cloud gateways including Amazon Bedrock and Microsoft Azure. IBM says its quantum computer completed a difficult 70-qubit test in 16.1 minutes while detecting errors and validating its results. Researchers working with Nvidia then reported a classical method using 256 H100 GPUs and projected that a comparable workload could be completed in 10.6 minutes. Also in this episode: Flock Safety faces 90 municipal cancellations in August; infostealer malware is hijacking authenticated Claude sessions; and New Brunswick introduces a bell-to-bell school phone ban, child-care screen limits and social-media blocking on school networks. CHAPTERS 00:00 Headlines and introduction 00:47 OpenAI cuts Cursor off before Astra 02:53 IBM's 70-qubit quantum test 04:13 Classical computers challenge IBM 05:34 Flock Safety backlash accelerates 07:53 Infostealers hijack Claude sessions 10:22 New Brunswick's school phone ban 13:08 Wrap-up and credits

AI Hardware Arms Race & Humanoid Robot Olympics: Apple Chips, OpenAI 'Jalapeno,' Meta Models, and Robot Prices Plummet Jim Love, Marcel Gagné, and retired John Pinard cover tech news and a deep dive on robots. They note Linux's 35th anniversary and how the name changed from "Freax" to Linux. They discuss Apple's AI-focused hardware advances (integrated CPU/GPU/neural processing and shared memory) enabling large local model inference, and OpenAI's "Jalapeno" inference chip built with Broadcom, developed in about nine months, delivering 1.5–1.9x more AI work per watt, raising questions for NVIDIA's CUDA moat. They touch on Meta's strong new Muse Glimmer/Spark models despite Reuters reporting failed internal AI rollout, morale issues, and metrics like 220% more AI-assisted coding with 40% reaching production alongside slower issue resolution. Bill Gates warns AI's growing capability could drive permanent job loss and dual-use security/bio risks. They then review China's World Humanoid Robot Competitions, rapid performance gains, autonomous team sports, and one-shot robot learning, plus falling robot price points—from $8K household task bots to Unitree's ~$5,900 humanoid and a ~$1,300 wheeled home helper—arguing autonomy is advancing fast. 00:00 Welcome and Introductions 00:48 Linux Turns 35 02:29 Apple Goes All In 06:18 OpenAI Builds Chips 10:41 Meta Models and Missteps 17:25 Why AI Rollouts Fail 20:33 Bill Gates Sounds Alarm 24:59 AGI and Persistent Agents 33:03 Humanoid Robot Olympics 34:56 Robot Sprint Breakdown 35:49 Humanoid Games Highlights 37:04 Soccer Bots And Teamwork 39:19 From Sports To Warfare 41:04 One Shot Robot Learning 46:33 Robots Making Robots 49:11 Home Robots Price Drop 52:14 MicroDuck And Cheap Bots 53:46 Unitree R1 Under 6K 58:55 Nori Budget House Helper 01:04:18 Autonomy And Price Point 01:05:55 Optimus 2027 Outlook 01:07:42 Weekly Wrap And Signoff

OpenAI has unveiled Jalapeño, its first custom AI inference chip, and the early results could challenge one of Nvidia's biggest advantages. OpenAI says Jalapeño delivers between 1.5 and 1.9 times more AI work per watt than the systems it tested against. SemiAnalysis watched benchmark runs inside OpenAI's lab and reported strong performance-per-watt results against Nvidia Blackwell. But the bigger story may be software: OpenAI is using Codex to write optimized kernels, potentially weakening the nearly 20-year CUDA software moat that has helped protect Nvidia from competitors. Also on this episode of Hashtag Trending: A Reuters investigation reveals how Meta's ambitious AI-driven restructuring plan fell apart. AI-assisted code increased dramatically, but gains in actual product features were far smaller, while technical incidents increased and employee morale deteriorated. Canada is investing more than $500 million in 64 new research chairs as it tries to turn decades of brain drain into brain gain. The real test will be whether Canada can keep the intellectual property, companies and economic benefits created by those researchers. And Twitter is back. Sort of. Operation Bluebird has launched a new social network using the Twitter name and the familiar language of tweets and retweets. Elon Musk's X says it still owns the trademarks and is suing. Hashtag Trending is hosted by Jim Love. 00:00 Headlines 00:27 OpenAI unveils the Jalapeño AI chip 01:39 Can OpenAI challenge Nvidia's CUDA moat 03:02 Meta's AI restructuring plan backfires 04:50 Canada makes a major brain-gain push 06:33 Can Canada turn research into Canadian industry 07:38 X battles a new company called Twitter 09:18 Wrap-up

Meta will pay at least $12.1 billion—and potentially as much as $17.1 billion over ten years—to settle claims that Facebook and Instagram harmed young users. Jim Love examines why the record settlement may still be a smart move for Meta. The company admits no wrongdoing, limits its legal exposure and avoids a verdict that thousands of other plaintiffs might have used in their own lawsuits. TIME also reports that Sam Altman expects OpenAI to have an internal system he would call artificial general intelligence by the end of 2026. What would qualify as AGI, and why would it matter more than passing the Turing Test? We separate confirmed reporting about OpenAI's delayed Astra models and cybersecurity concerns from unconfirmed rumours involving persistent memory and continual learning. Also in this episode: The Justice Department and FBI disable QScan and QTRouter, two hacking platforms allegedly used by Chinese state-sponsored attackers against NASA, the Federal Reserve, the U.S. Senate and hundreds of other organizations. The Environmental Protection Agency proposes removing federal minimum public-participation requirements for minor-source air permits, potentially allowing some data centre generator permits to be approved without public notice or comment. 00:00 Today's tech headlines 01:22 Meta's record teen-safety settlement 04:22 Altman says AGI could arrive this year 08:06 FBI disrupts China-linked hacking platforms 10:44 EPA proposes permit-notice changes 14:11 Wrap-up and where to listen Hashtag Trending for Thursday, August 27, 2026, hosted by Jim Love.

Apple is making a much bigger bet on local AI. Jim Love looks at the new M6 Mac mini, Apple's first 2nm processor, with Neural Accelerators built into every GPU core, dual 16-core Neural Engines and significantly faster processing. At the high end, the new M5 Ultra Mac Studio can offer up to 80 GPU cores and 512GB of unified memory, potentially allowing enormous AI models to run entirely on the Mac. The performance comes with a price. The new M6 Mac mini starts at $899 U.S., compared with the original $599 starting price of the M4 model. But Apple did announce one major price cut: its famous polishing cloth drops from $19 to $9. Also on Hashtag Trending: A new investigation warns that a catastrophic failure of the aging U.S. electrical grid could leave parts of the country without power for 18 months or more. Large transformers are often custom built, difficult to replace and increasingly hard to source. An earlier FERC analysis identified roughly 30 particularly critical substations and concluded that losing the right nine simultaneously could potentially trigger a nationwide blackout. Then, nearly 6,000 senior executives were asked what AI has actually done for their companies. More than 90 percent reported no impact on employment and 89 percent reported no improvement in labour productivity. Yet executives still expect AI to reduce employment. Separate research finds that AI-related layoffs can make employees less receptive to the very technology companies hope will increase productivity. And Linux turns 35. Linus Torvalds introduced his new operating system in 1991 by calling it "just a hobby" that wouldn't become big or professional. Today Linux powers servers, cloud infrastructure, Android and every one of the world's Top500 supercomputers. Hosted by Jim Love. 00:00 Headlines and intro 00:33 Apple M6 and the local AI push 03:45 Apple's $9 polishing cloth 04:24 America's aging power grid 06:06 AI productivity reality check 08:07 Linux turns 35 10:07 Wrap-up and where to listen