I’ve spent my career leading creative teams and building brand stories that connect from Hulu, T-Mobile, Samsung, Dignity Health, Dunkin’ and Jack in the Box. 

My work lives where emotion meets innovation, and lately, that means embracing the incredible power of Generative AI to reimagine what’s possible.

I believe in using AI as a creative amplifier, not a shortcut, empowering tools that free creatives to focus on the story, the craft, and the spark of the idea. Whether it’s scaling brand storytelling, launching employer brands, or helping a team unlock their next big idea, I love the moment when imagination becomes motion.

Alongside storytelling and design, I’m deeply experienced in performance creative - concepting, testing, and scaling campaigns built to drive measurable growth across social, digital, and video platforms. My data-driven approach ensures every creative decision connects back to outcomes, using insights to refine messaging, optimize formats, and increase conversion and engagement.

A few career highlights I’m proud of:

• T-Mobile – Led cross-functional teams across online and offline communications for the "Uncarrier", delivering engagement lifts of 30%+.

• Hulu – Helped reinvent campaign production for episodic employee stories that are "Powering Play," tripling creative output while cutting turnaround times nearly in half.

• Samsung – Brought the EVP story "Own the Unknown" to life with visuals and motion design, delivering 250+ formats in four weeks.

• Dunkin’ – Used influencer content and social vertical video to spark a 57% surge in applications.

• Jack in the Box – Built a dynamic content system powered by generative design and geo-targeting, driving 50M+ impressions and a same-store sales lift.

I’m passionate about blending storytelling, design, data, and emerging tech to create work that resonates…not just because it’s smart or scalable, but because it feels alive. And I love helping teams do the same: empowering creatives to explore, experiment, and create what’s next.

Core Skills: Creative Direction • Storytelling • Brand Strategy • Employer Branding • Performance Creative • Generative AI • Data-Driven Design • Creative Technology • Creative Ops • Leadership & Mentorship • Adobe Creative Cloud • Firefly • ChatGPT • Sora 2 • Higgsfield • Runway • Midjourney • ElevenLabs • Google Labs • Veo 3.1 • Canva • Kittl • HeyGen • Descript

Bruce Carey

There's a weird comfort in chaos. 🌀 I used to think I needed everything mapped out before I could start. 

The right brief, the right team, the right mood. But the best work I've been part of started with a mess. A half-baked idea. A deadline that felt impossible. A team that didn't quite fit together, at least not at first. 🤝

It's funny, the things you try to avoid are often the things that push you further. 🚀 I'm not saying I love the stress. I don't. B

ut I do love what happens when you stop waiting for perfect conditions and just start. Sometimes, the mess is the point. Or at least, it's where the good stuff hides. ✨ You know what I mean??

undefined undefined

AI Creative Director | Generative Design & Innovation Leader

There are days when leading a creative team feels like herding cats 😺. Other days, it's more like being the cat, curious, distracted, chasing after things no one else can see.

I used to think my job was to have all the answers. But now I know that more often, it's about asking the right questions, or just making space for someone else's eccentric idea to breathe 💡.

The best moments aren't when everything goes according to plan. They're when someone surprises you, or when the room gets quiet because everyone's thinking, really thinking, about what comes next 🤔.

It's messy. Sometimes it's frustrating.

If I'm honest, I still get nervous before big pitches. Still worry I'm missing something obvious. But maybe that's part of what keeps it interesting. You never really arrive. You just keep moving, keep creating and making...and keep learning how to listen 🚀.

I keep coming back to this idea: perfection is boring.

There’s something about a story with a rough edge, a design with a little friction, a campaign that doesn’t quite fit the mold. It sticks with you. It feels alive.

AI can polish anything until it gleams, but sometimes I want to see the brushstrokes. The fingerprints. The places where someone took a risk and it didn’t land perfectly, but it landed honestly.

Maybe that’s what we’re hungry for now. Not just seamless experiences, but ones that remind us there’s a human on the other side. Someone who tried, stumbled, and kept going.

I don’t know, maybe I’m just tired of the algorithmic shine. Or maybe I just miss the feeling of being surprised.

Either way, I’ll take a little imperfection over predictability any day.

Curiosity isn't a faucet you can turn on. Some days, it's flowing non-stop. 🔍 Other days, nada, no matter how hard you look...

In creative work, I've noticed that when curiosity fades, energy dips with it. The work becomes routine, ideas dry up, and even the best teams feel stuck.

But when curiosity sparks, even for a minute, everything shifts ⚡. A random question, a strange headline, an offhand comment in a meeting, suddenly, you're digging deeper, seeing opportunities you'd have missed.

You can't force curiosity. But you can make space for it. Pause when something feels odd. Let yourself follow weird questions. Encourage your team to share out-there ideas 💡.

It won't always work, and that's fine. The point is to follow where curiosity leads, even if it looks strange at first.

What's the last thing you got curious about at work? Did it take you anywhere unexpected? 🤔

Do you notice how perfect AI polish feels...so empty? ✨ I'll take rough edges any day. Doodles without the bezier curves. Songs with cracks in the voice. Photos a bit blurry. Stories told with real voices.

We're told AI will make everything seamless and flawless. But when things are too smooth, it feels like something's missing...Real connection comes from imperfection. From that human touch. That's what makes work feel alive. 🤝

So I'll keep choosing the messy, real parts over spotless perfection! 🎨

No alternative text description for this image

There's a weird tension in creative work right now. On one hand, we have more tools than ever. AI, automation, data that can predict what people want before they even know it themselves 🤖. On the other, there's this ache for something real. Something that doesn't feel like it was assembled by a committee of machines.

I catch myself toggling between the two. Chasing efficiency, then craving a little chaos ⚡. Wanting to impress, then wanting to connect 🤝.

It's exhausting, honestly. But maybe that's the point. Maybe the future isn't about choosing sides, but learning to live in the middle. To let the tech do its thing, but not let it drown out the real, wonderful, unpredictable parts of being human 🌱.

Some days I get it right. Some days I don't. But I'm starting to think the struggle is where the good stuff lives ✨.

I once lost a pitch because I refused to rush a half-baked idea. The client wanted something yesterday. I wanted something unforgettable for them, not slop churned out. 

We lost to an agency that promised speed. Three months later, their campaign fizzled out. No one remembered it. Not even the client.

Meanwhile, I kept tinkering with my team and staying in touch with the client on LinkedIn. My team scrapped drafts. We let the ideas breathe. We let it get weird. We let it get...good.

When we finally shared it, the client prospect was a little skeptical. It was different. But it was real. It made people stop scrolling 👀. It made them feel something 💡. That campaign went viral 🚀. Not because it was fast. Because it was right.

In a world obsessed with instant content, slow creativity feels almost rebellious 🕰️. But it is the only way to make work that matters. The only way to build something that lasts longer than a trending hashtag.

If you want to stand out, resist the urge to sprint all the time. Let your ideas simmer. Let them surprise you. Let them grow teeth. The best creative work is never microwaved. It is slow-cooked, with care and curiosity.

Do you have a story about an idea that was revived from the dead? Did it pay off?

Curiosity is a strange thing. It's supposed to be this endless well, but some days it just dries up. I'll sit in a meeting, nodding along, and realize I haven't asked a real question in ther meeting. Not out loud, anyway. 

It's easy to get comfortable, to let the routine do the thinking for you. But then something small breaks the pattern...a weird headline, a new voice in the room, a project that doesn't fit the mold. 🌀

Suddenly, I'm awake again. The questions come back, and with them, the urge to poke at things, to see what's underneath. 🔍

I wish I could bottle that feeling. Or at least remember how to find it when the days get heavy. 🌧️

Maybe it's about giving yourself permission to not know. To admit you're lost, or bored, or just plain stuck. That's usually when the good stuff starts to surface.

Curiosity isn't a given. It's something you have to fight for, especially when everything around you is pushing for certainty. I forget that more often than I'd like to admit. ⚡

Sometimes I wonder if we're all just pretending to know what we're doing.🤔 I've been in office rooms where everyone nods along, but you can feel the uncertainty hanging in the air. Nobody wants to be the first to say, I don't get it. Or, this isn't working.

But the older I get, the more I respect the people who are willing to admit when they're lost. Or when they're scared. 🗣️

Uncertainty isn't a weakness. It's the starting point for something new. 🌱 I wish I'd learned that sooner. I spent years trying to look like I had it all together. Now, I'm more interested in what happens when we drop the act. 🎭

It's not comfortable. But it's real. And sometimes, that's enough.

🍯🐻 Happy National Winnie the Pooh Day! 🐻🍯Celebrating the honey-loving bear who taught us about kindness, curiosity, friendship, and the power of a good snuggle. Pooh always said it best: “Sometimes the smallest things take up the most room in your heart.” 💛 Here’s to simple joys and big hearts today.
#WinnieThePoohDay #PoohBearWisdom #HeartwarmingMoments #SimpleJoys

🩳🔥 Happy International Sweatpants Day- get comfy y'all! 🔥🩳 Whether you’re channeling Rocky sprinting up the Philly steps in full training sweats 🥊💥…or The Dude sipping a White Russian in maximum-comfort mode 🕶️🥃… today is the day to celebrate the true hero of modern life: sweatpants.
Comfort. Confidence. Inspiration. Sweatpants contain multitudes. 😄 Honor the day (and night!). Get comfy! #InternationalSweatpantsDay #ComfortMode #RockyVibes #TheDudeAbides #LinkedInAfterDark

Curiosity is a strange thing. It's supposed to be this endless well, but sometimes it just... dries up. 🤔

I used to think I was naturally curious, but lately I've realized it's something you have to feed. Not with big, dramatic gestures, but with small acts. Reading something outside your comfort zone 📚. Asking a question you're a little embarrassed to ask. Letting yourself be a beginner again.

It's uncomfortable. But it's also where the best ideas hide. 💡

I wish I could say I do this every day. Truth is, I fall into ruts like anyone else. But when I remember to nudge myself out of autopilot, even just a little, things start to shift. 🔄

Maybe that's all it takes. A little discomfort, a little wonder, and the willingness to not know for a while. 🌱

Does everything online looks "too perfect" and doesn’t feel real anymore? 🤔 AI is cranking out polished content every minute. Our feeds are packed with flawless graphics, clever headlines, and data-driven stories. 

But sometimes it feels like the spark is missing.

The best creative moments aren’t engineered. They come from late-night pushes, team brainstorms, things that are messy and risky and eccentric as hell. 🎨

Tech moves fast, but real storytelling still starts with a human. AI can remix and polish, but it can’t replace that raw, honest touch...the part where you don’t know if your idea will land or flop. ⚡

It’s tempting to chase perfection. But the work that sticks with people is usually the one that’s a little rough around the edges. The pitch that made you sweat. The sketch that wasn't perfect. The idea you almost gave up on. 📝

In a sea of synthetic content, embracing a little honest mess might be the boldest thing you can do. 🚀

🧙‍♂️📚 Happy J. R. R. Tolkien Day! 📚🧙‍♂️ From Middle-earth to Mount Doom, Tolkien showed us the power of imagination, friendship, and courage...and that even the smallest person can change the course of the future. 🌋✨
This evening is perfect day to revisit "The Lord of the Rings" trilogy on streaming for the next 9 hours before work on Monday...my precious!!! #TolkienDay #MiddleEarth #MyPrecious #Storytelling #LinkedInAfterDark

I've been thinking about the word authentic. It gets thrown around so much it almost stops meaning anything.

But then I see a piece of work that feels true...raw, a little unpolished, and it hits me. That's what I'm hungry for. Not the perfect story, but the one that feels like someone actually lived it. 🎬

It's hard to make space for that kind of honesty. There's always a voice in the back of my head telling me to clean it up, make it safer, more universal. But the stuff that sticks with me, the stuff I remember, never plays it safe. 🔥

I guess I'm still learning how to let the rough edges show. To trust that people can handle the mess, maybe even need it. 🤔 It's a risk, every time. But I'd rather take that risk than make something forgettable. 🎯 Sometimes I wonder if I'm the only one who feels this way. Probably not... #BeTrue #Authenticity

Lately, I've been craving the old analog days. Letraset press type, Xerox machines, sketch pads, X-Acto blades... Not because I think AI tech is the enemy, far from it. 

I love what AI can do, the way it cracks open new possibilities 🤖. But there's something about the friction of making things by hand that I miss ✍️. The smudges, the crossed-out lines, the way you could see the work in the work.

Now, everything is so clean. So fast ⚡. You can generate a hundred options in seconds, but sometimes it feels like none of them really belong to you. Like the machine is doing the heavy lifting, and you're just curating the results.

I know that's not entirely fair. AI is a tool, not a replacement. But I worry about creators losing the fingerprints, the little imperfections that make something feel alive. Maybe that's nostalgia talking. Or maybe it's a reminder to slow down, to leave a little room for the mess. To let the process show.

I don't have a solution. Just a stubborn hope that we can keep the human in the loop, even as the tools get smarter. Maybe especially then ✨.

Technology is supposed to make things easier, but sometimes it just makes things...LOUDER. 🔊 There's so much noise out there. 

So many tools promising to streamline, optimize, automate. And yet, the work that sticks with me, the stuff that actually moves people...never feels like it came from a machine. It's rougher around the edges. Sometimes it's a little awkward. Sometimes it's a lot awkward. But it's always unmistakably human. 🤝

I think of AI as a collaborator, not a replacement. Like a really smart assistant who can fetch things, organize chaos, maybe even surprise you with a connection you hadn't seen. But it can't feel for you. It can't care. 

That's still our jobs. And honestly, I'm glad for that. Because if everything sounded the same, I'd probably stop listening. 

So here's to the weird, the imperfect, the stuff that doesn't quite fit. That's where the real stories live. 🌱✨ (At least, that's where I want to spend my time.)

Ever notice the best ideas show up when things get messy? 🎨 

Real stories, the kind people remember, rarely come from a flawless plan. They grow out of mistakes. Wrong turns. Half-ideas that nobody takes seriously at first. 💡 Some of my favorite work started as something awkward or unfinished. Sometimes the detour is the actual path.

Perfection is overrated. The details we can't control, the mess we try to avoid, that's what makes storytelling feel real and human. 🤝

Maybe we need to give ourselves (and our teams) more permission to let the process stay a little chaotic. 🔄 Because when you loosen your grip, that's when something interesting happens.

Your best ideas commonly show up when you're not even trying. 💡 You sit down, ready to be creative... and nothing. Blank page, empty head. 

But take a walk 🚶♂️ to Starbucks, watch some YouTube, play laser with my cats, and suddenly, BOOM, ideas everywhere...

I used to fight it, thought I was slacking off if I wasn't "being productive." Now I just let my mind wander and see what happens. Turns out, forcing creativity is the fastest way to kill it. ❌ 

So if you're stuck, don't stress. Go do something else for a while. Let your brain breathe!

🙏 Honoring Dr. Martin Luther King Jr. today: "Only in the darkness can you see the stars." , MLK 

Today is a reminder that even in challenging times, hope shines through people...through kindness, courage, and the belief that together we can build something better. 🤝 Here's to lifting each other up, choosing empathy, and continuing the work he championed. 💪❤️

Hi everyone,

If you would like to connect with me and we don't know each other PLEASE reach out to contact@thiswayglobal.com and share what you would like to talk about so you can be connected to the right person on my team. You will have far greater success than reaching out to me here in LinkedIn.

My team will accept public speaking requests, customer invites and investor communications here.

Otherwise, feel free to "follow" me 

Hope you have a great day!
Angela

I am the Founder/CEO of ThisWay Global, an international, VC-backed HR tech company that was incubated at ideaSpace – University of Cambridge, UK, with new offices now in the US. ThisWay's team is comprised of experienced entrepreneurs, global PhD’s in the areas of machine learning and AI, as well as talent from across the globe. 

Over the last twenty years, I have founded and built three companies, in sectors that range from architecture to the licensing industry. The common core in each has been my ability to build consensus amongst stakeholders, whether they be individuals, businesses or international governments.

I am a recognized thought leader on the topics of "Diversity and ROI" and "AI in the new world of work."

Without knowing my history, it would seem that the leap from land development to tech development is a big one. However, ThisWay is the manifestation of my drive to solve big problems. As early as 1992, when I founded the Texas A&M College of Architecture Career Fair, through my award-winning land development plan in the early 2000s, I have always sought to find meaningful solutions to big challenges.  

How we got to where we are today:
IBM Silver Business Partner
Google Accelerated Founder and Startup
Business Journal Finalist - Women of the Year 2023
ideaSpace-University of Cambridge: Alumni Founder
Institute of Enterprise and Entrepreneurs: Lifetime Fellow
Cambridge Wireless: Founder Member
Enterprise Nation: Entrepreneurship Mentor for both Growth and Hiring

Get in touch with me here or via either of the channels below:

Angela Hood

A genome-wide association study that normally takes months 

Finished in 20 minutes with Claude.

That's not a typo. 

Researchers at Stanford built Biomni, an AI agent that connects Claude to hundreds of databases and scientific tools. 

One test: data cleaning, analysis, statistical modeling, results. 

Twenty minutes.

Another project that would take a human expert three weeks finished in 35 minutes.

The bottleneck in scientific research has never been thinking. 
It's been data. Months spent cleaning messy datasets. Weeks parsing through literature. Grad students arguing over spreadsheets about which genes matter.

Claude is collapsing that part of the pipeline.

MIT's Cheeseman Lab uses Claude to interpret patterns from millions of CRISPR gene knockouts. The lab director can recall about 5,000 gene functions from memory, but Claude still catches connections he misses. Stanford's Lundberg Lab is using it to predict which genes matter based on molecular properties instead of guesswork.

Anthropic's AI for Science program gives free API credits to researchers working on high-impact projects. 

Claude for Life Sciences now connects directly to Benchling, PubMed, and 10x Genomics. 

They're basically removing every excuse to stay stuck on busywork.

It's only a matter of time before we see an AI-assisted breakthrough.  Fingers crossed for this year!

What do you think it will be?

AI for B2B expert/ Forbes 50/50 List / INC Magazine Founder / Google Accelerated / IBM Think Keynote / Outstanding Alum@TAMU & Founder/Alum Uni of Cambridge: ideaSpace Founder/Alumni

Microsoft is giving 12 months of Copilot and LinkedIn Premium Free 

To every college student.

The headline: "Democratizing AI literacy and preparing the workforce."

aka whoever owns the tool you learn on becomes the tool you lobby for at work.

Google did something similar last year with Gemini.  Plus think back to their 15-year campus strategy. Free Gmail, Docs, Drive for students. Now Google Workspace is the default in 90% of schools and a good chunk of enterprises. 

That didn't happen by accident.

Microsoft is not just handing out Copilot licenses. They're bundling LinkedIn Premium, launching global educator communities, creating new credentials, building AI-powered lesson planning tools. 

They're structuring the entire learning experience around their stack.

The free part is real. The value is real. Students and educators genuinely benefit.

Microsoft's betting that a $150 lifetime value of free subscriptions converts to years of enterprise spending down the road. 

Given their track record with enterprise bundling, that math probably works.

Google vs Microsoft continues on the college campus front.  Which stack is your school currently using?

ThisWay Global, maker of Amalgamy.ai, is now an approved OEM vendor with TD SYNNEX.

Amalgamy.ai is built for anyone selling to companies that operate data centers. 

This proprietary data and compute orchestration platform is platform-neutral and designed specifically for heterogeneous data center clusters. 

It removes the bottlenecks associated with data movement and scheduling, enabling more efficient AI, HPC and data center operations while also maintaining security. 

Because it's agnostic, it works across all data center environments, regardless of processor brand or type.

The ugly secret out there is the amount of wasted capacity and its impact on ROI.  How do you actually run that hardware at full capacity?  With Amalgamy.ai.

TD SYNNEX partners can now leverage the Amalgamy.ai software suite and appliance for all CPU/GPU/TPU, storage, and accelerator deployments. By perfecting the fusion of data and metal, data orchestration and scheduling bottlenecks will be solved in one motion.

If you're a TD SYNNEX reseller, Systems Integrator, or services partner working with customers investing in GPUs, accelerators, or high-performance storage,let’s deliver a high value solution together this quarter.

We're looking to collaborate on opportunities to fuse data and infrastructure together.

www.thiswayglobal.com

#TDSYNNEXUS #ThisWayGlobal #AmalgamyAI

ThisWay Global has had an incredible start in 2026 and we are hiring!
We will start posting the jobs here next week and on our company LI page.
As we are an AI and infrastructure acceleration company, all jobs are technical in nature. Expect open roles in these catagories:
#sales engineers
#software engineers
#sales execs with experience in software, data center, infrastructure and construction industries

We know there are a lot of people being layed off from the companies featured in this Crunchbase announcement. and we are fully remote, hiring US based employees.  So comment here to be sure you see all the jobs as they open up over the coming weeks.

Open article: Tech Layoffs: US Companies With Job Cuts In 2024 And 2025 by news.crunchbase.com

Tech Layoffs: US Companies With Job Cuts In 2024 And 2025

OpenAI just opened a request for proposals worth billions 

In domestic manufacturing contracts.

The deadline is June 2026. 

They're looking for US-based suppliers to build data center equipment, consumer electronics components, and robotics parts. 

They'll review rolling basis and fund the winners..

When the largest AI company in the world publicly commits to building supply chains domestically, other companies watch. Competitors follow. 

Investors fund the manufacturing startups that can compete. That's how momentum shifts.

OpenAI's Stargate initiative promised 10 gigawatts of AI compute capacity. 

They're already halfway there with announced projects creating "good jobs and local economic growth" across the country. 

Now they're formalizing that preference for domestic manufacturing across the entire supply chain. Not just chips, but racks, cooling systems, power electronics, testing and assembly.

Large-scale AI infrastructure doesn't have to be built offshore. It can be built here, faster, more resilient, and with domestic leverage.

Whether that actually happens at scale depends on whether US manufacturers can compete on cost and speed. 

Do you think domestic manufacturing can hit the price points needed to win against incumbent overseas suppliers, or will cost pressure force most contracts overseas anyway?

AWS turns 20 next month, and the party's getting expensive.

For years, cloud pricing felt inevitable,

Just another rising line on the spreadsheet. 

But 2026 flips that script. 

Amazon announced GPU pricing cuts up to 45% last year. Don't expect that again.

Energy costs are climbing. Data centers are power-hungry. Every major cloud provider (Amazon, Microsoft, Google) is dumping billions into AI infrastructure and training. 

They need revenue to fund it.

So who wins in 2026?

Organizations that double down on FinOps. 

The ones treating cloud spend like a business problem. Accurate tagging, reserved instances, spot pricing, renegotiating contracts. Those companies will carve out 10-20% savings and keep capacity while competitors absorb price hikes.

Everyone else gets squeezed.

The flip side: companies scrambling to justify cloud bills might also realize some workloads don't need premium cloud anymore. 

Neocloud providers (specialized AI-optimized environments) are undercutting conventional cloud on certain jobs. 

That's pressure AWS, Azure, and Google didn't have before.

The 20-year cloud story started with cheap compute changing everything. 

This chapter is about who can optimize when the easy savings are gone.

Are you in a position to absorb cloud cost increases, or are you already hunting for optimization levers?

New skills can get you a 15% raise. 

But companies are cutting entry-level jobs where people learn those skills. 

So who's actually going to get them?

Jobs that need new skills pay about 3% more. 

Stack four or more of those skills and you're at 15% premiums in the UK, 8.5% in the US. 

That extra money doesn't just sit there, it cycles through local economies when people spend it.

AI skills pay well, but occupations exposed to AI have seen 3.6% lower employment over five years. 

Companies using generative AI are making a clear choice: automate what entry-level workers used to do, hire fewer of them.

So now we've got a labor market that's splitting. High-skill workers are doing fine. Everyone in the middle doing routine work? That's where the squeeze is happening. 

And if you're young trying to get your foot in the door, there are fewer doors.

Finland and Denmark are putting real money into training programs. Other countries have mismatches, either tons of demand with nobody to fill it, or plenty of talent with nowhere to go.

It comes down to decisions about who gets access to training, how we handle competition, what safety nets look like, and whether education systems adapt.

Plus what are companies actually doing with the money they save from automation?

AI labs are basically grading their own homework. 

Miles Brundage spent seven years at OpenAI working on policy for AGI.

Now he's launching AVERI 

(AI Verification and Evaluation Research Institute) with $7.5M in funding and a blunt argument: 

Companies testing their own safety is insane. It's like your vacuum cleaner testing itself and promising it won't catch fire.

Here's how it works today. Companies conduct safety evaluations in-house. They publish whatever results they want. Nobody checks. Each lab decides what tests to run, how rigorous they should be, and what counts as safe.

Brundage and 30+ AI safety researchers published a research paper proposing "AI Assurance Levels," Which range from basic third-party testing all the way to "treaty grade" assurance for international agreements. 

They don't want to do audits themselves. They want to catalyze an industry of auditors from existing firms and new startups.

Who forces this? Insurance companies requiring audits before underwriting policies. 

Investors demanding verification before billion-dollar checks. 

The EU AI Act already hints at this direction for high-risk deployments.

But it's tricky to find auditors.  AI safety auditing requires rare skills and technical AI expertise plus governance knowledge. 

Everyone with that combination is getting lucrative offers from the companies they'd be auditing. 

Brundage's bet is building teams from audit firms, cybersecurity, AI safety nonprofits, and academia.

Dario Amodei says AI will replace every software developer within a year.

Demis Hassabis, who just won a Nobel Prize for his AI work, 

says current systems are "nowhere near" human-level intelligence.

Yann LeCun, the Turing Award winner who pioneered neural networks, goes further. 

He says LLMs will never achieve human-level intelligence. Not in 10 years, not in 50. Never.

At Davos this week, AI leaders couldn't agree on whether we're one year or one decade away from AGI.

Amodei claims we'll have Nobel-level AI scientists in two years and 50% of white-collar jobs will vanish in five. 

Sam Altman says we're already slipping past AGI toward superintelligence.

Meanwhile, Hassabis says we need "one or two more breakthroughs" and puts the timeline at 5 to 10 years. 

LeCun says the entire industry is "digging the same trench" with LLMs when we actually need completely different architectures.

LeCun's critique hits hard. If LLMs can't build world models or predict consequences, we won't get domestic robots or self-driving cars, let alone AGI. He left Meta over this exact disagreement.

The business world doesn't care. 

Cognizant estimates current AI could unlock $4.5 trillion in U.S. productivity. 

Trillions are flowing in regardless of who's right about AGI.

Does the actual timeline matter, or is the debate itself what makes the funding possible?

OpenAI is now predicting who's under 18 on ChatGPT. 

Based on behavior, account age, activity patterns, usage times. 

If the model thinks you're a teen, safeguards kick in automatically.

The safeguards make sense: no graphic violence, self-harm content, or material promoting extreme beauty standards. 

Parents get quiet hours, memory controls, distress alerts.

But here's the operational tension.

The model will get it wrong. Some adults will get flagged as teens. Some teens won't. When misclassification happens, you need a fast way to correct it. 

OpenAI built that, a selfie verification through Persona. 

Simple enough on paper.

But at scale, this becomes a support and trust problem.

If you're an adult locked into the teen experience, how fast can you actually get verified? What's the friction? Do you abandon the account or jump through hoops? If the system over-classifies, you're building resentment with users who have nothing to hide.

And if teens figure out the signals the model watches, activity times, account patterns. They'll just game it. 

The system becomes an arms race between OpenAI's refinements and user behavior modification.

The harder question for platforms: How do you manage this at millions of accounts? 

What's the cost of handling appeals and misclassifications? Does the support burden outweigh the safety benefit?

Age verification online is hard. Building the tech is one thing. Running it operationally is another.

How would you handle this if you were OpenAI's head of trust and safety?

Nvidia's $100 billion OpenAI investment was never actually promised.

Not quite. 

Jensen's walking back the September announcement by emphasizing it was "never a commitment." 

He's now saying Nvidia will invest "one step at a time" and that the current round won't approach $100 billion, but OpenAI invited them to invest up to that amount. 
 
Which makes sense we would immediately jump at the large figure and react. 

The Journal reported Huang had privately expressed doubts about OpenAI's spending discipline and questioned their business approach. 

"That's nonsense." In his own words.  

Nvidia needs OpenAI as a customer. OpenAI needs Nvidia's chips to exist. 

This strategic partnership will continue and I'm sure the actual investment numbers will still be substantial. 

With reported Q3 revenue of $57B, there's money to throw around here.

Gartner: 75% of firms will need a digital sovereignty plan.

But it's not about where your data lives. 

It's about control.

AI workloads are bringing this issue to a head. 

Storing data in a specific country was the first-order problem. Running proprietary models on that data, under a specific jurisdiction's laws, is a much harder one. 

For auditors and customers, simple data residency is no longer a sufficient answer.

We're seeing a clear shift in architecture. Sovereignty is moving from a bolted-on security control to a core design principle, embedded directly into the software stack. 

This looks like a customer-operated control plane, where identity, access, and encryption keys are kept strictly within the organization's boundary, even when running on a hyperscaler's hardware. 

The boundary becomes cryptographic and operational, not just geographic.

This changes the infrastructure discussion from choosing a cloud region to building a verifiable environment that you operate. F

or companies deploying proprietary AI against sensitive data, this kind of control is fast becoming a competitive requirement, not just a compliance checkbox.

ChatGPT started as a research preview. 

Three years later, it's the infrastructure layer for how people think.

That's OpenAI's framing in their new strategy note.

People didn't just download ChatGPT to experiment. They folded it into their lives. Students used it to untangle homework at midnight. 

Parents used it to plan trips and manage budgets. Writers broke through blank pages. Then they brought it to work.

At first in small ways. A draft refined before a meeting. A spreadsheet checked one more time. A customer email rewritten to land the right tone. 

Then it became part of daily workflows. Engineers reasoned through code faster. Marketers shaped campaigns with sharper insight. Managers prepared for hard conversations with better context.

What started as a curiosity tool became infrastructure.

Here's the architecture OpenAI is describing: 
compute scales at 3x per year 
(0.2 GW in 2023, 0.6 GW in 2024, 1.9 GW in 2025). 

Revenue scales at the exact same curve 
(10x in three years, $2B to $20B). 

Stronger models unlock better products. Better products drive adoption. 

Adoption funds the next wave of compute. The flywheel compounds.

The business model is deliberate too. Consumer subscriptions for individuals. 

Usage-based pricing for teams where costs scale with real work. APIs for developers. Free tiers for adoption. Commerce and advertising for decision-making moments. 

Every monetization path is calibrated to the value delivered.

Compute is still the scarcest resource. Three years ago OpenAI relied on a single provider. Now they're diversified across an ecosystem, training frontier models on premium hardware for capability, serving workloads on cheaper infrastructure for efficiency.

The vision for 2026 is "practical adoption." 

Close the gap between what's possible and what's actually being used. Health, science, enterprise. Places where better intelligence becomes better outcomes.

Does OpenAI continue to grow the adoption curve it needs to sustain and thrive?

Neoclouds are tracking toward $180 billion in revenue by 2030.

You probably haven't heard of them. 

But Nvidia's channel chief calls them the "next phase of opportunity" for the AI data center boom.

So what are neoclouds?

They're smaller, AI-specialized cloud providers built specifically for training and inference workloads. Think CoreWeave and Nebius. 

They compete with AWS, Azure, and Google Cloud, but they're laser-focused on AI infrastructure instead of general-purpose computing.

Right now they're pulling in $23 billion annually. If the projections hold, that becomes $180 billion by 2030.

That's not niche territory. That's a legitimate market segment carving out space from the hyperscalers.

Nvidia's Craig Weinstein says this is an "ever-growing and important segment" for the broader ecosystem. 

World Wide Technology, ranked #9 in the Solution Provider 500, is already generating substantial revenue from hardware resale, data center design, and managed services for neoclouds.

If you're training massive models, you don't need AWS's 200+ services. You need optimized GPU clusters, fast interconnects, and teams that understand AI workloads.

We've spent 15 years watching everything consolidate into three hyperscalers. 

Now AI workloads are specialized enough to fragment the market again.

If you're building or optimizing data center infrastructure for AI workloads, let's talk about solving orchestration bottlenecks together.

#AmalgamyAI

Google just funded the guy who co-authored "Attention Is All You Need."

Llion Jones helped invent transformers. 

Now he's "absolutely sick of them."

Google announced a partnership with Sakana AI, backing the Tokyo startup with capital and access to Gemini and Gemma models. 

Sakana isn't building bigger transformers. 
They're exploring fundamentally different architectures inspired by biology.

Their Continuous Thought Machines incorporate timing at the neuron level. Something transformers completely ignore. When solving mazes, the model's attention literally traces the path like a human thinking step-by-step.

Their AI Scientist automatically generated a research paper that passed peer review at ICLR. The first fully AI-authored paper to do it.

Last month, their ALE-Agent became the first AI to win a major programming optimization contest, beating 804 human competitors.

Why does Google care? 

Because Deepmind CEO Demis Hassabis just said at Davos that AGI probably needs "one or two more breakthroughs," specifically around continual learning and memory. 

Those are exactly what Sakana researches.

Of the eight transformer co-authors, only two still work at major frontier labs. 

The rest scattered to startups and other labs. Jones is the most vocal about moving beyond what they built.

Sakana will actually still use Gemini extensively.

But is this Google admitting the transformer paradigm might be incomplete?  

Hassabis said language models could be "a key component" of AGI rather than "the only component." 

Google is paying to keep a seat at the table if something else wins.

What happens when AI stops waiting for your question, 

And starts investigating on its own?

OpenAI Prism embeds GPT-5.2 directly into your research paper. 

It reads your entire manuscript: structure, equations, citations, figures. Like a co-author who's actually read the whole thing. Then it pulls relevant papers, converts your whiteboard sketches into LaTeX, and reasons through your equations in context. 

Kevin Weil at OpenAI said: "2026 will be for AI and science what 2025 was for AI and software engineering."

That's basically Claude Code for scientists.

Google's Agentic Vision does something different. When AI looks at a photo now, it doesn't just glance and guess. It actually investigates. Zooms in, crops sections, checks details. Accuracy jumps 5-10%. 

A startup using it for architectural reviews got 5% gains on code compliance checking. That's real money.

Then there's Moonshot's Kimi K2.5. One prompt generated a 100MB Excel storyboard with 55 consistent scenes for a film. 

The model figured out what needed to happen, split the work between specialist agents, and stitched it back together. 

Their video-to-code feature is wild: record your screen while browsing, upload the video, it clones the entire site with all the interactions.

As AI applications advance, it will stop waiting for you to ask the right question. 

It's now asking what it needs to know, what tools to use, how to check its own work.

Investigates, coordinates, verifies.

We're about to see more and more workflows change to human in the loop management of AI agent outputs. 

How would that affect your job?

Goodbye text to video. 

Hello world model? 

Google releases Project Genie, which is being labeled as a world model. You describe an environment in text, upload a sketch, pick how you move through it, and the AI generates the entire world in real time as you explore it. 

Walking, flying, driving. Doesn't matter. The world builds itself around your input.

Right now it's limited to 60-second sequences. Grainy visuals. 

But that's not the point.

For decades, 3D world building has required artists, level designers, physics programmers. 

You want a procedurally generated dungeon? You need someone to code it. You want a city layout? An artist draws it. A game environment? A whole team builds it.

Genie flips that. You describe what you want. It generates it. Then it simulates it. In real time.
The implications split two ways.

For game devs and architects, this is a productivity tool. Rapid prototyping. 
Sketch an idea, iterate instantly, move to polish. That's useful.

But for the people whose job is "turn a description into an interactive space," this is a different conversation.

Google's being cautious. It's locked behind their Ultra subscription, $250 a month. Gradual rollout. Not exactly screaming from the rooftops about it.

The shift from "humans build worlds" to "AI generates worlds on demand" doesn't feel like it's coming in 5 years. It feels like it's here. 

Which industry adopts Genie first: games, architecture, education, or something else?

OpenAI just turned coding into management.

You're now the PM of your robot employees.

The Codex macOS app lets you spin up multiple AI agents working on 
different parts of your codebase at once.

Each agent gets its own git worktree, so they work in parallel without colliding. You switch between them, review changes, and merge when ready. It's more about coordination than writing code faster.

Skills are bundled instructions that connect Codex to external tools and complete full workflows.

- Deploy to Vercel.
- Triage bugs in Linear.
- Generate images for game assets.
- Create documents.

OpenAI built an entire Mario Kart clone using a single prompt and 7 million tokens, costing $175 in API fees.

Backing up claims that the unit of work may change.

Then there are Automations, which run tasks on a schedule. Daily issue triage. Finding CI failures. Generating release notes. The work your team forgets until it becomes a crisis is now delegated to a background agent.

Some early access users claim, "I feel more like a manager than ever, but I also feel way more productive than ever."

Managing an agent squad has been the dream for a while now.

This gets us one step closer. 

Is anyone making the switch from Codex / Claude Code?

Cisco surveyed 5,200 security professionals across 12 markets. 

90% expanded their privacy programs because of AI. 

93% plan to spend more this year.

Privacy spending has tripled. 

38% now drop at least $5 million annually on it, up from 14% last year.

These privacy frameworks are unlocking faster AI deployment. 
And more importantly, better customer trust, an actual competitive advantage.

But a lot of companies are stuck in the middle.
65% can't access quality data efficiently. 
85% say data localization requirements are adding cost and complexity. 

They're spending millions on governance while still fighting basic data hygiene.

Jen Yokoyama at Cisco put it this way: "Organizations must deeply understand and structure their data to ensure every automated decision is explainable. It's not just for compliance, but a necessary scaling engine for AI innovation."

Companies with mature governance can move fast. 

The others are paying what I'd call the AI tax. 

Higher costs, slower rollouts, fragmented systems.

Would love to hear what you think of this technology!

Board & Executive Advisor on AI Accountability and Human Impact Former Fortune 500 finance executive helping leaders make responsible decisions before risk, regulation or reputation force the issue

A college student with quadriplegia types math equations with her tongue.

No surgery. No headgear. Just a retainer.

She’s studying math and computer science. She takes notes in class, texts friends, and researches assignments—often in places where voice control isn’t an option. Libraries. Lectures. Group study rooms.

The device is MouthPad^, built by MIT spin-off Augmental. It’s a custom 3D-printed retainer that sits on the roof of your mouth and turns tongue movement into cursor control.

Think about what assistive tech usually asks people to accept:
 ↳ Surgical implants, recovery time, and real risk
 ↳ Visible headgear that turns you into “the person with the device”
 ↳ Voice commands that fail the moment the room needs to be quiet
 ↳ Price tags that put independence out of reach

What Augmental built instead:
 ↳ A touchpad for the tongue: swipe to move, press to click
 ↳ Head-tracking for broad motion, tongue reserved for precise control
 ↳ Bluetooth pairing like a normal mouse—no special software
 ↳ $1,500 during early access
 ↳ Users wearing it up to 9 hours a day with low fatigue

Here’s the part that stopped me.

Her mother called receiving the MouthPad^ “the most significant moment since her injury.” 

Not the diagnosis. 

Not the rehab milestones. 

A dental retainer that gave her daughter control back—without making her life louder, more visible, or more medical.

In a lecture hall, no one knows she’s controlling her laptop. 
That’s not a side benefit. 
That’s the design.

For boards and executive teams, this isn’t a feel-good edge case. It’s governance. When non-invasive, low-cost accessibility exists and organisations still choose systems that shut people out, the issue stops being “technology limits” and becomes leadership. Procurement standards, inclusion policy, and digital decisions determine who gets full access—and who is left working around choices made for them. That’s when accessibility stops being a design preference and becomes an accountability issue.

And it raises a blunt question:

Why did we spend decades trying to restore digital access by opening skulls when a workable interface might already be inside the mouth?

♻️ Share if you think accessibility shouldn’t require surgery—just better design.

This is one of the stories from my keynote on Designing Accessible Futures: AI, Ethics and Inclusion in Practice for boards and leadership teams.

Source: MIT News and Augmental (MouthPad^ assistive technology).

Video Insta luminica.ai

taplio

The Battle

Bruce Carey

Angela Hood

Followers

Posts

Engagements

Likes

Performance ScoreBruce Carey's LinkedIn Performance

Performance ScoreAngela Hood's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Bruce Carey

Angela Hood

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Bruce Carey's LinkedIn Performance

Performance ScoreAngela Hood's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreBruce Carey's LinkedIn Performance