Samsung Electronics Brings ChatGPT and Codex to Employees
OpenAI is calling this one of its largest enterprise AI rollouts ever — Samsung Electronics is deploying ChatGPT Enterprise and Codex to employees worldwide. This is the same Samsung that famously banned ChatGPT after a source code leak in 2023, which makes this either a remarkable turnaround in corporate risk tolerance or a very expensive act of forgetting. Either way, when a chaebol this size goes all-in on OpenAI tooling, it’s a signal the enterprise market is past the pilot phase.
Beyond Siri: Here Are the Practical AI Features Coming to Your iPhone in iOS 27
Siri grabbed the WWDC headlines, but TechCrunch is digging into the quieter AI features threaded throughout iOS 27 — and there are more of them than most people noticed. Apple’s strategy here is clearly “AI everywhere, but make it feel like it was always there,” which is either brilliant product design or the most expensive stealth campaign in consumer tech history. The real test isn’t the demo; it’s whether any of this still works six months after launch without a major patch cycle.
The New Siri Makes One of Apple’s Most Convenient OS Features a Cumbersome Mess
The Register’s counter-programming to Apple’s WWDC glow: the new Siri has apparently turned Spotlight — one of macOS and iOS’s most quietly beloved features — into something that feels like being forced to talk to a chatbot just to open a file. “Distressingly like Google AI Overviews” is a comparison that should make any product manager lose sleep. The irony of building a smarter assistant that makes the overall experience dumber is a very 2026 problem.
Salesforce Rolls Out New Slackbot AI Agent as It Battles Microsoft and Google in Workplace AI
Salesforce has rebuilt Slackbot from the ground up — out with the notification pinger, in with the “fully powered AI agent” that can search enterprise data, draft documents, and take actions on your behalf. The timing is pointed: Microsoft’s Copilot is making identical promises, and Google’s Workspace AI isn’t standing still. The workplace AI war is now less about features and more about whose agent you trust with your calendar and your inbox, which is a genuinely interesting trust problem nobody’s fully solved yet.
No Longer Just a Copilot, Microsoft’s AI Wants to Take the Wheel
Microsoft’s always-on agent evolution is getting The Register treatment, and the headline says everything: “provided you trust it with practically everything.” That’s the quiet ask underneath all the productivity promises — hand over your emails, your calendar, your documents, your workflows, and in exchange you get… faster meetings. The value proposition isn’t nothing, but the trust transfer is enormous, and most enterprise IT departments are still working out what “agentic AI” means for their security posture.
When the Trump Administration Cracks Down on Anthropic, Who Benefits?
TechCrunch’s Equity podcast is asking the right question here — because in regulatory crackdowns, there’s always a winner, and it’s rarely the public. If the administration keeps tightening the screws on Anthropic, the most obvious beneficiaries are OpenAI and Google, neither of which needs the competitive help. The deeper issue is whether AI safety concerns are driving this or whether it’s the usual Washington sport of picking favorites under the cover of national interest.
Sick and Wrong: Ontario Auditors Find Doctors’ AI Note Takers Routinely Blow Basic Facts
This one keeps circulating for good reason: Ontario auditors found that 60% of evaluated AI medical scribe systems mixed up prescribed drugs in patient notes. Sixty percent. These aren’t edge cases or hallucination quirks in a chatbot answering trivia — these are medication errors baked into clinical documentation. The AI-in-healthcare enthusiasm is real and sometimes warranted, but this is a sharp reminder that “good enough for most cases” is a dangerous bar when the failure mode is a wrong drug on a chart.
Listen Labs Raises $69M After Viral Billboard Hiring Stunt to Scale AI Customer Interviews
Listen Labs put a billboard in San Francisco displaying what looked like random gibberish — actually AI tokens that, decoded, contained a job offer — spent $5,000 doing it, and parlayed the resulting attention into a $69 million raise. I’ll grant them this: in a funding environment where you’re competing against Mark Zuckerberg waving $100 million at engineers, a clever puzzle-billboard is a better ROI than a Super Bowl ad. The product itself — AI-powered customer interviews — is less flashy than the origin story, but the fundraising meta-game here is genuinely impressive.
Bottom Line
Whether it’s Samsung betting the enterprise on ChatGPT, every workplace app elbowing its way into your workflow, or medical AI getting drug names wrong 60% of the time, today’s theme is the same: AI is no longer knocking at the door — it’s already inside, and we’re still figuring out which rooms it should have access to.