
88% resolved. 22% stayed loyal. What went wrong?
That's the AI paradox hiding in your CX stack. Tickets close. Customers leave. And most teams don't see it coming because they're measuring the wrong things.
Efficiency metrics look great on paper. Handle time down. Containment rate up. But customer loyalty? That's a different story — and it's one your current dashboards probably aren't telling you.
Gladly's 2026 Customer Expectations Report surveyed thousands of real consumers to find out exactly where AI-powered service breaks trust, and what separates the platforms that drive retention from the ones that quietly erode it.
If you're architecting the CX stack, this is the data you need to build it right. Not just fast. Not just cheap. Built to last.
Hi friends!
This community is growing, and I want to get to know the brilliant minds behind the screens. I’ve put together a super quick survey to learn more about your roles, your industries, and what truly matters to you. Got a spare minute? Your feedback means the world!
Today:
A Sneak Peek at Anthropic’s Most Powerful AI Yet
DeepMind Alumni’s Robotics Startup 'Physical Intelligence' Eyes a Staggering $11 Billion Valuation
Google Prepares to Bankroll a Multibillion-Dollar Texas Data Center for Anthropic
Samsung Turbocharges Its Internet Browser With Perplexity’s AI Engine
Google Shakes Up Global Communication With Real-Time Headphone Translation

It turns out even the most advanced AI companies in the world are vulnerable to simple human error. Anthropic accidentally left a draft blog post sitting in an unsecured content management system, unintentionally revealing the existence of a highly secretive, unreleased model called "Claude Mythos."
The leaked data suggests Mythos represents a massive "step change" in capabilities, particularly in coding, academic reasoning, and cybersecurity. In fact, the model is reportedly so far ahead in its cyber capabilities that the leak actually sent cybersecurity stocks tumbling due to the unprecedented risks it might pose. It’s a true blockbuster reveal, completely by accident.
Behind all these magical models are incredibly heavy, expensive physical operations. In a massive infrastructure play, Google is nearing a deal to help finance a multibillion-dollar data center down in Texas.
This new facility will be leased directly to Anthropic. While it might sound like standard corporate real estate news, it's actually a huge deal. It deepens the strategic and financial relationship between the two companies as the physical race for AI compute power continues to heat up. To build the best models, you need the biggest data centers, and Google is making sure Anthropic has the hardware to keep pushing the envelope.
Physical Intelligence, a robotics startup founded by former Google DeepMind researchers, is currently in talks to raise $1 billion.
If successful, this deal would push their valuation to an eye-watering $11 billion. That would double their valuation in just four months. The team is building advanced, general-purpose AI models designed to power robots to perform a huge variety of complex, physical tasks. Investors are clearly betting massive amounts of capital that the next big frontier is bringing AI out of the screen and into the physical world.
🧠RESEARCH
PixelSmile is a new AI system for editing facial expressions in photos. Normally, changing a smile without accidentally altering the person's identity is difficult. This tool perfectly separates expression from identity. It lets users adjust smile intensity smoothly and accurately, keeping the original face natural while making the smile shine.
Voxtral TTS is a new text-to-speech AI that clones voices using just three seconds of audio. It creates highly realistic, emotional speech across nine languages. By combining different methods to process sound and meaning, it sounds far more natural than older systems, making digital voices incredibly lifelike and expressive.
LagerNVS is an AI tool that instantly generates new camera angles of a scene from flat photos. Instead of slowly building a complex 3D model first, it relies on hidden 3D knowledge learned from past data. This clever shortcut allows it to render high-quality, completely new viewpoints in real time.
📲SOCIAL MEDIA
🗞️MORE NEWS
Perplexity on Samsung Devices Samsung added a smart tool created by Perplexity to its internet browsers on phones and computers. This helper allows users to easily sum up long websites and quickly search through their past internet visits. It shows how smart computer programs are moving beyond basic chat to actively helping people finish online tasks.
Google Live Translate with Headphones Google brought its instant translation tool to Apple iPhones and expanded it to more countries around the world. Users can wear normal headphones to hear over 70 foreign languages translated into their own language without any delay. The software keeps the original speaker's tone and rhythm, which makes conversations between different cultures feel much more natural.
The Decadelong Feud Shaping the Future of AI A ten-year personal fight between the leaders of two major tech companies, OpenAI and Anthropic, is guiding the future of the entire industry. The conflict began over deep disagreements about product safety and whether to focus on the public good or just making money. These intense arguments caused a group of workers to leave and start their own rival business, which is now worth billions of dollars.
OpenAI Surpasses $100 Million Annualized Revenue OpenAI is now making money at a pace of $100 million a year from showing advertisements inside its popular ChatGPT service. They reached this major financial goal just six weeks after starting a small trial program to test out the ads. The company plans to keep growing by launching a simple system where any business can log in and buy ad space directly.
Midjourney Profitable, Chasing Hardware Dreams Midjourney, an independent company that makes art from typed words, is now highly profitable, meaning it makes much more money than it spends. Despite facing giant rivals like Google, the company survives by keeping its team small and letting users create images together in public chat rooms. To keep growing, they are now hiring engineers to build physical computer devices instead of just software.



