Perplexity's AI Glow-Up

The Street

The Productivity Glow-Up AI Needed

Today, we’re merging our usual How to AI section with the main feature to cover Perplexity’s new AI tool that not only puts the internet’s knowledge at your fingertips, it puts the knowledge to work.

Which is what we all want at the end of the day: AI that works for us. Because while Creating Studio Ghibli images is cool, its utility is short-lived.

🔥AI That Actually Gets Stuff Done

If you love building or bringing ideas to life, there’s never been a better time to be a creator. AI continues to blaze past the hype phase and to become a real creative partner. And we’re not talking about another LLM wrapper here, but real, additional value.

Quick Recap: What’s Perplexity?
Perplexity, founded in 2022, is like Google—but with citations, real answers, and none of the SEO clutter. And for a while, it wasn’t clear how Perplexity would compete as Google rolled out Gemini (their own AI powered search feature). But instead of doubling down on search, Perplexity started building something better.

Maybe that’s why they’re reportedly closing another $500M funding round at a $14B valuation?

🚀 Meet Perplexity Labs
Just yesterday, Perplexity launched Labs, transforming it from search tool into a full-blown AI project studio. Whether you're building a pitch deck or launching your next side hustle, Labs drastically reduces the time and effort it takes to bring your idea to life. Just you, a good prompt, and you’ll have a working output in minutes.

Here are the core features:

• ✅ Code execution

• 🌐 Web browsing

• 🖼️ Image generation

• 🔁 Multi-step workflows

• 💾 Persistent memory

• 🧩 Build and deploy basic web apps within your project

That last bit is key. What good is analysis if you can’t display and share it? Labs lets you spin up dashboards, slide decks, and websites with no external dev tools.

All this costs $20/month, which isn’t free, but is pretty reasonable compared to the $200 Pro tiers out there (looking at you, ChatGPT).

📽️ Watch the demo

🤔 The Real Test: Can AI Startups Survive?

While Labs feels like a step forward, it also raises a bigger question: what makes an AI startup truly durable?

In a recent piece titled “99% of AI Startups Will Be Dead by 2026,” author and podcaster Srinivas Rao lays out a sharp argument. His point: Many AI startups are just thin layers on someone else’s model—with nothing defensible. That’s why they’re called LLM wrappers—at best, fragile businesses; at worst, features waiting to be absorbed by the platforms they depend on.

Here’s his TL;DR:

• 💀 Most AI startups just repackage ChatGPT (or another LLM)

• 💸 They charge for things you could do yourself on said LLM

• 🧱 They precariously rely on expensive rented infrastructure (Microsoft, NVIDIA)

In a sea of flashy LLM wrappers, substance matters. Perplexity Labs is a glimpse of what AI can become when it moves beyond answers and starts delivering real, usable output. But in a crowded market, hype fades fast.

The tools that last won’t just be impressive—they’ll be useful. Consistently.

AI built for the day-to-day of running a business

If you’re running a small business and trying to scale without burning out, HoneyBook can quietly take over a lot of the details that eat your time.

It’s free to try, and worth checking out if you’re looking for a real productivity boost!

Get the tools, gain a teammate

• Impress clients with online proposals, contracts, and payments.

• Simplify your workload, workflow, and workweek with AI.

• Get the behind-the-scenes business partner you deserve.

Try HoneyBook for free

Heard in the Server Room

Anthropic CEO Sees AI Jobpocolypse In Next 5 Years

Anthropic CEO Dario Amodei says the AI jobpocalypse isn’t a distant sci-fi plot—it’s arriving fast. He predicts up to 50% of entry-level jobs could vanish and unemployment could surge to 10–20% within five years, as smarter-than-college-level bots move into white-collar roles like coding, finance, and consulting. While DC is busy blocking China and boosting AI adoption (cue the “One Big Beautiful Bill”), Amodei is ringing the alarm—business leaders are quietly prepping for mass automation, and the public’s not ready.

Nvidia’s Winning Streak Continues With Stellar Earnings

Nvidia crushed earnings again, raking in $44B last quarter—up 69% YoY—thanks to sky-high demand for its AI chips, which now power 88% of its revenue. Even with a $4.5B hit from U.S. export bans on China sales, its data center biz (aka the AI engine) grew 73%. Nvidia's performance signals that global AI infrastructure spending—especially by giants like Microsoft and OpenAI—is only accelerating. As CEO Jensen Huang put it, demand is “incredibly strong,” indicating that the AI boom is still booming.

NYT Chooses Love Not War, Signs AI Licensing Deal With Amazon

The New York Times just signed a multi-year deal with Amazon to license its content—including NYT Cooking and The Athletic—for use across Alexa and Amazon’s AI models. It’s part of a growing trend: media outlets opting for licensing deals over courtroom brawls. While the Times is still suing OpenAI and Microsoft for copyright infringement, this Amazon pact gives it a new revenue stream as the company beefs up its generative AI lineup (like Alexa+, Nova models, and Bedrock).

Your Coworkers: Bob, Lisa, and GPT-4

Nearly 80% of businesses are already using AI to save time and cut down on busywork. But how it’s adopted isn’t exactly universal—some companies are automating everything, while others are still figuring out how to unmute on Zoom.

A recent study on AI usage revealed some striking trends:

📊 Key Stats Worth Knowing

• 78% of global companies are currently using AI. After years of plateauing, adoption jumped from 50% in 2022 to 78% in 2025, driven largely by the rise of our favorite generative AI tools.

• Large companies are 2x more likely to use AI than small ones—which is interesting considering it’s the smaller teams that could really use the extra help.

• India leads in adoption (59%), followed closely by UAE and Singapore, while the U.S. lags at 33%.

• Most common uses across companies include customer service (56%), cybersecurity (51%), and digital assistants (47%).

• The AI market is projected to grow to a whopping $1.85 trillion by 2030.

• Many AI startups remain small—83% have fewer than 10 employees (which makes sense when you realize some of those “team members” might be running on GPU).

• AI is pulling its weight—saving the average employee 2.5 hours a day, or roughly one long meeting that could’ve been an email.