😸 OpenAI's state of enterprise AI report 👀

Welcome, humans.

Mondays, am I right? Current mood coming back to work after spending all weekend vibe-coding with agents = about the same as this Tesla bot without an operator…

A HUGE thank you to everyone who subscribed to the YT channel this weekend! We had a massive spike in subs, and we’re just ~250 of y’all away from crossing 10K!

Y’all are simply the best and we couldn’t do any of this fun stuff without you!

LATER THIS WEEK: On Wednesday, we're going live with Connor Zwick, Co-founder & CEO of Speak, the AI language tutor that's teaching fluency the way humans actually learn: by speaking out loud.

Click this image, and on YouTube, click “Notify Me” to get notified when we go live!

We’ll chat with Connor and the AI live to test it out in real time. Plus, we hear they have some news coming soon… so tune in for that, too!

Here’s what happened in AI today:

OpenAI released a “state of the enterprise” report.
Trump says he’ll sign a “one rule” executive order to preempt state AI laws.
OpenAI and Instacart will let you check out groceries in ChatGPT.
Google told big advertisers it plans to run ads in Gemini in 2026.

Advertise in The Neuron here

OpenAI Just Revealed Who's Actually Winning at AI (And the Gap Is Massive)

Ever wonder if your company is using AI the “right“ way? Turns out most organizations aren't even close.

OpenAI just dropped its first-ever “State of Enterprise AI“ report, analyzing usage data from 1M business customers and surveying 9K workers. The findings are wild, and if you're not among the “frontier“ users, you're leaving serious productivity on the table.

The headline numbers are staggering:

ChatGPT now serves 800M weekly users, with Enterprise adoption exploding 9x year-over-year.
Weekly messages grew 8x since last November…
…And companies are consuming 320x more reasoning tokens than they were 12 months ago (makes sense, thinking is about only about 16 months old).

Now here's where it gets interesting. The average Enterprise worker saves 40-60 minutes daily using AI. Heavy users? They're saving more than 10 hours per week. Basically an extra workday per week!

Real-world examples:

Lowe's answers 1M customer questions monthly with Mylow, and when shoppers engage with it, conversion rates more than double.
Indeed reports job seekers using their AI Career Scout are 38% more likely to get hired and find jobs 7x faster.
Moderna compressed analytical work from weeks to hours in drug development planning.
BBVA built 4K+ Custom GPTs and automated 9K legal queries annually, redeploying three full-time employees to higher-value work.
Intercom's Fin Voice resolves 53% of support calls without human intervention—and when calls do escalate, they're resolved 40% faster.

The productivity secret? Frontier workers do AI different.

They send 6x more messages than the median employee.
Among data analysts specifically, top performers use the data analysis tool 16x more than their peers.
Workers who save 10+ hours weekly consume 8x more AI intelligence than those saving zero hours.

Even wilder: 75% of workers report completing tasks they previously couldn't perform at all, like coding, data analysis, or creating custom workflows.

Coding-related messages from non-technical workers jumped 36% as people discover they can now build things without traditional programming skills.

Btw, if you need a project idea to build on your newfound Codex skills, try building a Chrome extension! They’re super easy and can get up and running in 3 easy steps. Just say “let’s turn this into idea a chrome extension [idea here]. Walk me through it step by step.” ).

But there's a massive gap emerging.

Frontier companies send 2x more messages per employee than median firms, and 7x more through structured workflows like Custom GPTs.
Among monthly active Enterprise users, 19% have never tried data analysis features, 14% never used reasoning models, and 12% never tried search capabilities.
For those powerful capabilities to just sit unused? That's leaving money on the table.

Why this matters: OpenAI says organizations mastering AI deployment are pulling ahead, w/ 1.7x revenue growth and 3.6x greater shareholder returns (according to BCG). The gap between frontier firms and median performers will likely widen as AI capabilities improve faster than most companies can adapt.

If your team isn't among the frontier users yet, the good news is the playbook is clear:

Get executive buy-in
Enable data connectors.
Build repeatable workflows
Train people to use AI across multiple task types (forget trad domains; everyone should learn to code and use agents. It’s almost 2026, y’all).

The companies doing this now are building advantages that will be hard to catch as the pace of change gets faster and faster.

FROM OUR PARTNERS

Stop Clicking. Start Closing — Delve’s AI Does Your Questionnaires

Ever had to fill out a gnarly security questionnaire inside a vendor’s portal and hated your life?

Delve just launched an AI browser agent that does it for you.

It finds answers from your policies, past questionnaires, and docs, then fills everything in while you click next. Hours → minutes.Switch to Delve for compliance and use this - and every other gamechanging AI features at one simple cost.

If you’re using another platform for compliance, Delve will migrate you over or free.

The Neuron readers get $1,000 off compliance using the link below.

Set it and forget it >>> sweating questionnaires.

Book your demo now

Prompt Tip of the Day

Steal this mini-prompt template from Reddit and tweak it for your own work:

Context first. Start with: “Here’s the context: [paste it].”
Then the task. “Your task: [what you want], prioritized for [who/what matters most].”
Then constraints + role. “Act like [role]. Keep it under [length], give [N] options, match this structure: …”
Force questions. “Before you start, review everything and ask any clarifying questions you need. Number them and make them yes/no where possible.”
Add a correction loop. After the first answer: “Now critique your response for missing context, shaky assumptions, or fluff, and rewrite it with those fixes.”

P.S: We want to start sharing longer-form prompt advice, but we have limited space in the newsletter. So we’re bringing back the month’s prompt tip of the day archives, this time with longer-form pieces you’ll only find on the digest. Check it!

Treats to Try

*Asterisk = from our partners (only the first one!). Advertise to 600K readers here!

*The Artificial Intelligence of Things (AIoT) is driving efficiency, resilience and smarter decisions. Here’s five recommendations to accelerate AIoT value.
Speechify expanded its Chrome extension beyond text-to-speech by adding voice typing and a built-in voice assistant that can answer questions as you browse.
Cofounder connects to your work tools (Gmail, Slack, Notion, Linear) and automates workflows using plain English commands (like monitoring Linear for completed issues and updating your Notion release notes automatically).
Ripplica lets you record yourself doing a browser-based workflow once, like applying to jobs, triaging DMs, or pulling analytics, and then replays and adapts it with context-aware agents that click through any web app or legacy system on a schedule, even when you’re offline.
Magiclip turns long YouTube videos into viral Shorts and TikToks by letting you paste a URL and auto-clipping, resizing, subtitling, and even generating Veo 3-powered AI footage, images, and voices so you can crank out monetizable vertical videos in a few clicks—plans start around €15–19/month for 30 videos.

Around the Horn

This is the coolest visualization of an AI model brain I’ve EVER seen

Microsoft will remove its Copilot chatbot from WhatsApp on January 15 because the messaging app is banning general-purpose AI bots, so users will need to switch to Copilot’s own apps or the web to keep using it.
U.S. President Trump announced plans to sign a “one rule“ executive order this week seeking to centralize AI oversight and override state-level AI regulations.
OpenAI and Instacart partnered up so shoppers can plan recipes, build a grocery list, and check out from within ChatGPT, turning the chatbot into a full grocery-shopping agent that calls Instacart’s APIs under the hood.
IBM agreed to acquire data infrastructure company Confluent for $11B in cash to strengthen its enterprise AI and cloud offerings.
Google told major ad buyers it plans to roll out ads inside its Gemini chatbot in 2026, the first time it has directly briefed advertisers on monetizing Gemini.
Tesla demonstrated its Optimus humanoid robot running smoothly in lab tests, marking significant progress in bipedal locomotion with analysis suggesting speeds around 8.5 mph.
Medtronic received FDA clearance for its Hugo robotic-assisted surgery system for urologic procedures, challenging Intuitive Surgical's US market dominance
P.S: if anybody working at Apple is reading this atm, Justin Bieber has some UX design feedback for you.

FROM OUR PARTNERS

Model-driven Intelligent Agents

Strands Agents empowers you to create intelligent agentic systems that leverage reasoning models to make dynamic decisions and adapt to unexpected situations.

Just announced: support for TypeScript, improved support for edge devices, powerful evaluations, and a steering feature to help guide agents toward desired outcomes.

Start Building

Monday Meme

A Cat’s Commentary

ICYMI, this Pixar-style lamp is what they’re talking about (Ongo site / Interaction Labs site).

That’s all for today!

What'd you think of today's email?

Update your email preferences or unsubscribe here

3343 Perimeter Hill Dr. Ste. #100
Nashville, Tennessee 37211, United States of America