In partnership with
Welcome back to the Neural Net! It’s officially spooky season, but please refrain from any foolish wand-waving while you read the latest eerie AI news 👻
In today’s edition: The Neural Net’s guide to surviving the AI craze at work, an AI actress so good it’s spooking Hollywood, Sora 2 sets its sights on TikTok, and more.
▼
The Street
note: stock data as of last market close
▼
🤯 So Your Boss Has AI Fever: The Employee Survival Guide
I work in AI, so I know what it really takes to make it work. Which is why I get how overwhelming it feels when management buys ChatGPT licenses, sends an all-hands email—“AI is the future, go build agents!”—and suddenly you’re expected to be Tony Stark.
The reality? Gallup found only 8% of U.S. employees use AI daily (up from 4% last year). Another 19% use it a few times a week, while 60% barely touch it. And yet, more companies are starting to measure job performance by AI usage.
Two Roads to AI: Compliance vs. Transformation
As Microsoft CEO Satya Nadella put it, “The hardest part of AI isn’t the tech — it’s getting people to change how they work.” Two companies are testing very different ways of tackling that challenge:
1. Microsoft’s mandate model
Evaluated employees on Copilot and other AI tools, tying AI use to performance reviews
Mandated employees to adopt AI tools and incorporate them in their day-to-day
Message: AI isn’t optional, it’s required, and how good you are with it determines your career trajectory
2. Moderna’s evangelist model
Employees were trained and coached on how to use AI
Teams were encouraged to build real, problem-solving tools that work for them
Results: Moderna hit near-total AI adoption in just six months, with employees creating 750+ custom GPTs in two months, and averaged 120 ChatGPT conversations per user, per week.
👉 But here’s the thing: most of us aren’t setting company-wide policy; we’re living under it. So whether you’re living under the evangelist or mandate model, the next question is: how do you thrive (or at minimum, survive) as an employee when your company goes AI-crazy?
The Employee Playbook for Top-Down AI Fever
Start small: Pick one narrow task and prove how AI saves time. Quick wins earn credibility.
Translate hype into reality: When your boss says “make an agent,” pause. Half the time, they don’t really know what an agent is, and depending on your tools, it may not even be possible. Scope it down to a pilot that solves a real problem.
And remember: most bosses think a custom GPT is an agent. It’s not; it’s just ChatGPT with extra instructions. Agents can actually plan tasks and take actions. (But if your boss can’t tell the difference, a custom GPT might still do the trick 😉)
Don’t chase shiny objects: Every week a new tool pops up. Stick with what solves real problems, not what demos best.
Educate yourself—and others: Most people say “AI” and mean “chatbot.” Knowing there are actually different kinds of AI for different problems is how you go from tourist to local.
Ask for training: If you’re expected to use AI daily, push for support; it shouldn’t be on you to figure it all out alone.
Document experiments: Track what worked, what didn’t, and how long it took. Data gives you cover when expectations run wild.
Share, don’t compete: Instead of racing coworkers to “use AI the most,” swap tips and make adoption a team effort. 🤝
Set boundaries: Use AI where it helps, not everywhere. The wrong use just adds clutter and headaches. It’s up to you to experiment and define the boundaries!
Remember, if your boss has AI fever, it’s probably because their boss has AI fever. Even the CEO answers to the board. The reality is that good managers want their people to succeed, and they’re figuring this out just like you. And if they’re really struggling, sign them up for the Neural Net—they’ll be the next Steve Jobs in no time ;)
▼
In Partnership With Enterpret
How Canva, Perplexity and Notion turn feedback chaos into actionable customer intelligence
Support tickets, reviews, and survey responses pile up faster than you can read.
Enterpret unifies all feedback, auto-tags themes, and ties insights to revenue, CSAT, and NPS, helping product teams find high-impact opportunities.
→ Canva: created VoC dashboards that aligned all teams on top issues.
→ Perplexity: set up an AI agent that caught revenue‑impacting issues, cutting diagnosis time by hours.
→ Notion: generated monthly user insights reports 70% faster.
Stop manually tagging feedback in spreadsheets. Keep all customer interactions in one hub and turn them into clear priorities that drive roadmap, retention, and revenue.
▼
Heard in the Server Room
Perplexity just took its AI browser Comet worldwide—gratis (that means free, my fellow commoners). Originally only available for those on the $200/month subscription tier, Comet promises to be your digital sidekick: search, shop, draft emails, juggle tabs, the works. The move heats up the AI browser wars, with Google, OpenAI, and Anthropic all rolling out their own versions. And how do you win early adopters? Make it free, of course!
Tilly Norwood, an AI “actress,” has thrown Hollywood into meltdown after her creators revealed talks with a talent agency. With Instagram headshots and spoof clips, she looks like any other starlet, but the Hollywood writer’s union and celebs like Emily Blunt are calling her a fraud that “steals performances.” Creator Eline Van der Velden insists Tilly’s not replacing humans, but is just “a piece of art.” Still, the outrage shows nothing rattles Hollywood’s A-list like a digital brunette willing to work for less than $30M. Art is pain after all.
Apple is halting development of a cheaper version of the highly successful highly innovative highly useful highly expensive Vision Pro headset to take on Meta’s Ray-Bans smart glasses. The $3,499 headset hasn’t exactly flown off shelves, so Apple’s now cooking up two pairs of specs it hopes will: one that tethers to your iPhone (possibly next year) and another with its own display (targeted for 2028).
▼
In Partnership With Go-to-Millions
Your Boss Will Think You’re an Ecom Genius
Optimizing for growth? Go-to-Millions is Ari Murray’s ecommerce newsletter packed with proven tactics, creative that converts, and real operator insights—from product strategy to paid media. No mushy strategy. Just what’s working. Subscribe free for weekly ideas that drive revenue.
▼
✨ How to AI: OpenAI Bets Big on Sora 2 as a TikTok for AI Content
OpenAI just launched Sora 2, its upgraded video and audio generator, along with a new invite-only Sora app on iOS. Here’s what’s new:
More realistic video — smoother motion and better physics than Sora 1
AI speech generation — characters can now talk
Cameos — maybe the internet’s favorite feature: insert yourself or friends into clips, backed by a one-time ID check to keep deepfakes in check
Social app features — personalized feed, remix tools, and sharing... stepping straight into TikTok’s lane
Safety guardrails — watermarks, parental controls, and content checks
For now, access is invite-only in the US and Canada. That’s partly to manage hype, but mostly because generating photorealistic video is insanely compute-hungry.
👉The bigger play isn’t just better video; it’s OpenAI trying to build a social platform that feels a lot like TikTok, but fueled entirely by AI-made content. If it takes off, Sora could shift how people consume short-form video, give creators a new playground, and position OpenAI less as an API provider and more as a consumer-facing media company.
▼
That’s it for today! Have a spooky weekend, we’ll be back Tuesday.