🚨Disney’s OpenAI partnership is only exclusive for one year—then it can work with other AI platforms, signaling an early hedge against lock-in. Content giants want optionality as model capabilities (and pricing) keep shifting.

Transform your team’s static training materials into dynamic, engaging video guides with Guidde.

Here’s what you’ll love about Guidde:

1️⃣ Easy to Create: Turn PDFs or manuals into stunning video tutorials with a single click.
2️⃣ Easy to Update: Update video content in seconds to keep your training materials relevant.
3️⃣ Easy to Localize: Generate multilingual guides to ensure accessibility for global teams.

Empower your teammates with interactive learning.

And the best part? The browser extension is 100% free.

Check out Guidde

Most ads fail for one boring reason: the copy sounds like the marketer wrote it.

Not the customer.

And yeah, you can “do research” and read a few reviews and call it a day. Or you can do the thing that feels slightly illegal (it’s not) and walk away with 50+ lines of phrasing your market is already repeating to each other when nobody is selling them anything.

What’s happening

There’s a dead-simple loop that works in basically any niche. It looks like this.

You search Reddit for the exact pain point plus a modifier like “help,” “frustrated,” or “struggling.”

Grab 20 threads. Pull every comment. Dump it all into one doc.

Then upload the doc to Gemini and ask it to pull out the phrases people keep repeating and the beliefs hiding underneath them.

Also: what they’re mad about, in their words, without you cleaning it up.

You’re not asking for ad copy. You’re asking for the raw DNA -- how they frame the problem when there’s no brand voice, no funnel, no “positioning.”

Why this matters

Customers don’t describe their problems the way you do.

They don’t say “I need a solution for X.” They say “I’m so tired of ___” and then they add the small humiliating detail they’d never put in a survey. Reddit is where that detail shows up.

And once you have the language, your job changes. You’re no longer inventing messaging. You’re selecting the best lines, putting them in a clean order, and attaching your offer to what’s already true in their head.

That’s the whole game.

The part people miss

The output you want isn’t a list of “pain points.” It’s the repeated sentences.

The recurring metaphors.

The little beliefs that show up in ten different threads:

“Nothing works for me.”

“I’ve tried everything.”

“I don’t trust experts anymore.”

“If it was easy I would’ve fixed it already.”

That’s copy. That’s positioning. That’s objection handling. And it’s sitting there for free.

What this means

If you run ads, build a “language bank” the same way you build a swipe file. Make it market-first, not brand-first.

If you’re writing for a client, stop arguing about tone and start arguing about whether the headline uses the same nouns and verbs your customers use when they’re mad.

If you’re building a product, these threads are feature requests dressed up as venting. People will tell you exactly what they wish existed, they’ll even tell you what they tried that disappointed them and why.

What's next

The people who win with ads over the next year won’t be the ones with the best “creative strategy.”

It’ll be the ones who can pull real voice-of-customer language at scale, fast, and keep refreshing it as the market shifts. The internet changes weekly. Your messaging should too.

BTW: if you do this once, you’ll never go back to writing ads from a blank page. It feels like cheating.

Have an ad concept ready but don't want to deal with expensive shoots or stock footage? ScriptKit lets you generate, curate, and edit AI ads in one platform.

What ScriptKit gives you

Give ScriptKit a shot — go from concept to finished AI ad without wrangling teams or gear.

The Recap: Ford is launching a battery energy storage business aimed at powering data centers and supporting the grid, backed by about $2B in planned investment over the next two years. The company expects systems to start shipping in 2027, and it’s targeting about 20 GWh of annual storage capacity by late 2027 with 5 MWh+ LFP-based systems.

Unpacked:

Bottom line: AI is turning “car companies” into energy infrastructure companies. Ford’s bet isn’t just about batteries—it’s about owning a slice of the physical bottlenecks (power quality, peak shaving, resilience) that determine how fast new AI compute can come online.

VCs say consumer AI apps won’t stick without new hardware platforms, arguing software alone won’t create the next shift.

Nvidia acquires SchedMD (Slurm) and ships open models, tightening its grip on the AI infrastructure control plane.

Creative Commons backs “pay-to-crawl” marketplaces, pushing a new economic model for training data access.

Lightspeed raises a record $9B fund, signaling VC consolidation around capital-intensive AI bets.

Chai Discovery hits a $1.3B valuation building drug discovery foundation models—AI moving from text into molecular programming.

Ford adds a gas generator to its next F-150 Lightning, a market reality check on pure-EV strategy at the high end.

Merriam-Webster names “slop” word of the year, reflecting mainstream recognition of AI-generated content pollution.

iRobot invented the Roomba category but lost the future to players with stronger AI and ecosystems—hardware needs platforms to endure.

Best of AI™ Team

Was this email forwarded to you? Sign up here.