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.

See the data

Investors are watching Nvidia closely ahead of its annual GTC AI conference on March 16, where CEO Jensen Huang typically unveils new hardware and AI infrastructure plans. Nvidia has transformed from a gaming chip company into the backbone of the AI boom, with 91% of its revenue now tied to AI products. The question for investors is simple. At current valuations, is the next wave of AI announcements already priced in?

• Nvidia generated about $215B in revenue, with roughly 91% coming from AI-related products.

• The company’s GPUs remain the core hardware powering AI training and inference across the industry.

• Nvidia stock trades around 21× forward earnings, which some investors see as expensive.

• Analysts remain bullish overall. The average price target is $262, about 43% upside from current levels.

• CEO Jensen Huang’s GTC keynote historically moves markets, often revealing new chips, roadmaps, or ecosystem expansions.

Nvidia is no longer just a chip company. It is the infrastructure layer of the AI economy.
Every breakthrough in AI still flows through GPUs somewhere in the stack.
If compute demand keeps accelerating, Nvidia’s biggest risk is not competition. It is becoming the equivalent of the electricity grid for AI.

Stop thinking about AI purely as software and start mapping the hardware dependencies inside your AI stack.

Companies that secure long term GPU capacity and inference access will move faster than competitors still waiting in compute queues.

If Nvidia expands its platform at GTC, expect the biggest opportunities to appear in the software and services ecosystem that forms around its hardware moat.

Source: Yahoo Finance

AMI Labs, a new AI research company co-founded by Turing Award winner Yann LeCun, has raised $1.03B at a $3.5B valuation to pursue a different vision of AI. Instead of large language models, AMI is building “world models” that learn from real-world environments rather than text alone. The lab will focus on fundamental research and may take years before commercial products emerge.

• AMI Labs raised $1.03B to develop AI that understands the real world, not just language.

• The company is built around JEPA (Joint Embedding Predictive Architecture), a framework proposed by LeCun as an alternative to LLMs.

• The startup expects years of research before commercial products, prioritizing fundamental science over quick revenue.

• Major investors include **Jeff Bezos’ Bezos Expeditions, Eric Schmidt, Mark Cuban, and Tim Berners-Lee.

• The first application area may be healthcare, with digital health startup Nabla acting as an early partner.

Large language models are hitting reliability limits, especially in high-stakes fields like healthcare. World models represent a bet that the next generation of AI will learn from environments rather than text. If that shift works, it could redefine what counts as “intelligence” in AI and make today’s chatbots look like an early prototype.

Start investing in real-world data pipelines, not just text data. The next wave of AI may depend on sensor data, simulations, and physical environments.

Explore partnerships with research-first AI labs early. These ecosystems often define the standards the rest of the industry later adopts.

And rethink your AI roadmap. If world models succeed, the most valuable AI companies will not just process language. They will understand reality.

Source: TechCrunch

Perplexity has launched Personal Computer, a desktop AI agent that can directly access your files and apps to complete tasks. Unlike typical chatbots, the system acts more like a digital operator, executing actions on your machine rather than just generating text. The move brings agent-style automation, similar to OpenClaw, into a more consumer-friendly interface.

• Perplexity Personal Computer is a desktop AI agent that can open files, run apps, and complete tasks on your machine.

• Instead of commands, users give objectives like “create a podcast” or “build an educational guide.”

• The system operates locally but can be controlled remotely from any device.

• Perplexity says it includes safeguards such as user approvals, audit logs, and a kill switch.

• The company also launched an Enterprise version of its agent platform with corporate app connectors and APIs.

The AI interface is shifting from chat to action. Once agents can operate software directly, traditional apps become back-end infrastructure rather than front-end tools.
The companies that control the agent layer will quietly become the new operating systems of work.

Start designing workflows assuming AI agents will operate your software instead of humans clicking buttons.

Invest in APIs and structured data access because agents cannot automate systems that are locked behind messy interfaces.

And begin building internal governance for agent permissions and audit trails, because the companies that master AI delegation will outpace those still treating AI like a chatbot.

Source: Ars Technica

There are never slow weeks in AI. That’s for sure. There’s so much to see and so much yet to discover. You never know what comibination of words or letters may get unexpected results. Or whatever type of orchestration may get faster or more intelligent gains. There’s an entire world out there that has yet to be explored.

Soon there will be interview sections, expert opinions, and much more.

But for now, thank you for subscribing!

Till next time,

Edward Frank Morris
The Context_

Please note that all images are AI Generated and are for illustration purposes only.