THE UNIVERSAL AI CREDIT SYSTEM NOBODY'S BUILT YET
- candyandgrim
- Dec 16, 2025
- 13 min read
And why it's a £10 billion opportunity hiding in plain sight
A £10 billion opportunity is hiding in plain sight, and I'm genuinely surprised no one has built it yet.
As a Creative Director who's worked across motion graphics, 3D design, and AI creative tools for 15+ years, I've watched the explosion of specialized AI generation platforms with a mix of excitement and frustration. We now have hundreds of incredible tools—Kinetix for animation, FOSSA Tether for face rigging, Higgsfield for camera control, Pika Labs for region editing—each solving specific creative problems brilliantly.
But here's the problem: nobody can afford to subscribe to all of them.
And more importantly, nobody should have to.
The subscription chaos problem
Think of it like this: You wouldn't buy a different electricity provider for each appliance in your house. Yet that's exactly what we're doing with AI tools.
I'm a multi-discipline creative working in multi-discipline teams. Kinetix and FOSSA Tether are both in my top 10 gen-AI tools this year. But I don't use them every day—sometimes not even every week. So I'm constantly subscribing and unsubscribing, which is a total pain.
If Kinetix offered a one-time purchase (say £50-300 depending on whether it's a one-off tool or a suite) or a low annual tool license (£20-150/year) that I could use with a separate universal credit subscription? Take my money, please.
But currently, that doesn't exist.
Instead, we have:
With the exception of ComfyUI (which requires a beast of a machine for local AI generation), every platform requires credits. But no one is signing up for 10, 20, 50, 100 separate subscriptions per month.
It's like living in a world where every streaming service requires its own TV, its own remote, and its own cable box. Exhausting.
The aggregator problem: they solved half the puzzle
Now, you might be thinking: "But Simon, what about aggregator platforms?"
There are now many platforms that bundle multiple AI models under one subscription. Think of them as the Netflix of AI—one login, many shows:
Text/chat AI aggregators:
Magai - 50+ AI models including GPT-5, Claude, Gemini, DeepSeek
Poe (by Quora) - $19.99/month for GPT-4.5, Claude 3.7, DeepSeek
CosmicUp - $16/month for 30+ models including ChatGPT, Claude, Gemini, Midjourney
Mammouth AI - €10-60/month for text, image, and audio generation
TeamAI - From $5/user/month for enterprise teams
Abacus AI - $10/user/month for 20+ models
Creative-focused aggregators (including limited third-party generators):
Weavy (now acquired by Figma as Figma Weave) - Node-based platform combining Flux, Veo 3, Runway Gen-4, Imagen 3, Kling. From $19/month
Flora - $16/month+, infinite canvas with 50+ AI models including GPT-5, Gemini 2.5, Stable Diffusion 3.5, Kling 2.5, Pika
Luvian.ai - Complete creative suite with unified canvas for image, video, audio, text, and 3D generation
Recraft - Proprietary Recraft V3 model for brand-consistent design and vector generation. 5M+ users
ComfyUI (cloud versions) - Node-based local/cloud hybrid platforms
Ecosystem-specific credits:
Adobe Firefly - Generative credits across Adobe products
Google AI Credits - Works across Flow and Whisk
Microsoft AI Credits - For Microsoft 365/Power Platform
These platforms are like getting a universal remote—better than nothing, but they have fundamental limitations:
The chat aggregators give you access to language models, but you're using their interface, not the specialized native platforms. It's like watching all your shows on one streaming box, but you can only watch them in standard definition with limited features.
The creative aggregators (Weavy, Flora, Luvian, Recraft) are getting closer—they bundle generation models with professional editing tools in unified canvases. But you're still locked into their workflow and interface. It's better than juggling tabs, but you lose the specialized features that make individual platforms powerful.
The walled gardens (Adobe, Google, Microsoft) only work within their own ecosystems. It's like having a universal remote that only works for Sony products.
What if I specifically want to use Kinetix's character motion control AND FOSSA Tether's face rigging AND Higgsfield's camera angles AND Pika Labs' region editing? All need credits. All are isolated. All require separate subscriptions.
The aggregators solved one problem (multiple tabs and subscriptions) but missed the bigger opportunity: letting you use native specialized tools with universal credits.
The stock service evolution: a blueprint for what's next
To understand why a universal credit system makes perfect sense, let's look at how creative subscription services evolved. It's a story told in four acts:
Act 1 (1990-2010): the physical era
Creative assets lived in physical stock books and CD/DVD libraries. Need a photo? Order a catalogue, find the image, call to license it. Pay £5-50 per asset. It was like going to a video rental store every time you needed footage.
Act 2 (2010-2020): the Netflix moment
The industry shifted to unlimited download subscription models. Envato Elements, Adobe Stock, Motion Array—one monthly fee for unlimited access to millions of assets.
This was transformative. Instead of budgeting per asset, studios could budget per creator. Need a photo? Download it. Need music? Download it. Need templates? Download them. All included.
It was like going from pay-per-view to Netflix Unlimited.
Act 3 (2020-present): the hybrid catalogue
Adobe Stock now includes a mixture of gen-AI and human-generated content. The infrastructure exists to license both types under unified credits. It's like Netflix adding both licensed content and original productions—one subscription, diverse sources.
Act 4 (not yet built): universal AI generation credits
This is the next logical evolution—but nobody's built it yet.
Just as Envato Elements unified access to photos, video, music, templates, graphics, and fonts under one subscription, someone should unify access to specialized AI generation platforms under universal credits.
Act 5 (future): unlimited universal AI generation credits
This is the final stage
Runway does offer unlimited credits for its higher tier users, but these are not available in platforms where Runway is aggregated.
The infrastructure exists. The business model is proven. The demand is screaming loud.
So why hasn't anyone built it?
What should exist: the full vision
Here's what a proper Universal AI Credit system looks like—and it's more sophisticated than you might think:
Layer 1: the credit infrastructure (core)
A unified credit service that works across specialized creative AI platforms.
Not another chat aggregator. Not a walled garden ecosystem. A true credit infrastructure layer—think Stripe for AI generation instead of Stripe for payments.
You would:
Buy tool licenses separately based on their scope: Single-feature tools (like FOSSA Tether): £25-75 one-time Multi-feature platforms (like Kinetix): £100-300 one-time or £50-150/year Full creative suites: £300-500 one-time or £150-300/year
Subscribe to universal credits (see tiers below)
Credits work across ALL partner platforms seamlessly
It's like having one electricity account that powers every appliance in your house, regardless of manufacturer.
Layer 2: choose your cloud infrastructure
But here's where it gets interesting—and where nobody's thinking yet:
What if you could choose HOW your generations are processed?
Imagine a system where users select:
Option A: mixed cloud (optimal performance) The system automatically routes to the fastest/most capable cloud for each task. Pika generation? Runs on their optimized infrastructure. Kinetix animation? Routes to their servers. Like having a smart power grid that sources from the best generator for each appliance.
Option B: cheapest cloud Route all generations through the most cost-effective infrastructure. Maybe that's AWS bulk processing, maybe it's newer providers with cheaper rates. You get slower processing but stretch your credits further. Like choosing off-peak electricity rates.
Option C: greenest cloud Route through carbon-neutral or renewable-powered infrastructure. Google's renewable data centers, Microsoft's carbon-negative servers. You might pay a 10-15% premium, but your creative work has minimal environmental impact. Like choosing a green energy provider.
Option D: brand preference (Nvidia vs. Google vs. AMD) Some studios trust Nvidia's GPU infrastructure. Others prefer Google's TPUs. Let users pick their preferred compute brand, like choosing between Shell and BP—same fuel, different brand loyalty.
Layer 3: local gen-AI interpolation aggregator (ComfyUI gen 2?)
And here's the truly visionary piece nobody's discussing:
What if you could run generations locally when you have the hardware, and cloud-based when you don't?
Think of ComfyUI, but evolved. A system that:
✅ Detects your local hardware capabilities Got a Mac Studio with M3 Ultra? Run certain generations locally for free (zero credits).
✅ Intelligently routes between local and cloud Simple image generation? Local. Complex video synthesis? Cloud. Like having solar panels that supplement grid power—you use local when you can, cloud when you need it.
✅ Syncs workflows across devices Start a generation on your workstation, continue on your laptop, finalize on cloud infrastructure. Like having your music library sync across all your devices.
✅ Offers hybrid processing Use local hardware for initial passes (drafts, tests, iterations), then cloud credits for final high-quality renders. Like printing drafts at home but going to a professional printer for final output.
This solves the biggest problem with ComfyUI: it's powerful but requires expensive hardware. A hybrid system democratizes access—students can use cloud credits while professionals maximize their local hardware investment.
Why stock services are sleeping on this
I'm genuinely surprised that Getty, Envato, Adobe Stock, iStock, Shutterstock, ArtGrid, or Stockblock haven't seen this opportunity.
They already have:
✅ B2B relationships with creative agencies and studios
✅ Enterprise billing infrastructure
✅ Legal/indemnification frameworks proven through Act 2 (unlimited subscription transition)
✅ Brand trust for corporate buyers
✅ Experience managing cloud infrastructure costs
They're like taxi companies in 2009—perfectly positioned to build Uber, but missing the moment.
Three reasons they're missing this:
1. They're still fighting AI Getty sued Stability AI. Shutterstock cautiously partnered with OpenAI. They see AI as a threat to their revenue streams, not as their next business model. It's like Blockbuster seeing Netflix as competition instead of their own future.
2. They're thinking vertically instead of horizontally They want to own the full stack (images, video, music) rather than thinking as infrastructure providers for everyone else's tools. They're thinking "content marketplace" when they should be thinking "credit infrastructure."
3. They haven't connected the dots They already executed Act 2 (unlimited subscriptions). They're watching Act 3 (hybrid human/AI content). But they haven't realized Act 4 is staring them in the face.
It's like owning a massive electricity grid but refusing to let third-party appliance manufacturers plug in.
The business model that works
Here's how a universal AI credit system should work:
Tiered subscription + scalable infrastructure choice
Solo lite - £29/month
5,000 universal credits (good for image generation, testing workflows)
No indemnification (personal use only, at your own risk)
Access to 50+ partner platforms
Choice of cloud routing (mixed/cheapest/greenest)
Local generation support (hybrid processing)
Pro single user - £79/month
20,000 universal credits (covers serious image work + modest video projects)
Basic indemnification available as add-on (+£20/month for personal commercial work)
Priority support queue
Advanced analytics dashboard
Pro+ single user - £149/month
30,000 universal credits (multiple video projects per month, extensive generation)
Basic indemnification available as add-on (+£30/month for client work up to £50K projects)
Priority routing (faster processing)
Hardware optimization recommendations
API access for workflow automation
Team - £99/month per seat OR £499/month for 10-user shared pool
50,000 credits per subscription seat OR 500,000 shared credits (unlimited users accessing shared pool)
Medium indemnification available as add-on (+£50/month per team for client work up to £250K projects)
Team management dashboard (usage tracking, allocation controls)
Shared workflows and templates
Team training resources
Enterprise - custom pricing
Unlimited credits OR custom allocation pools based on team size
Full indemnification included (underwritten legal coverage for client work, unlimited project value)
Dedicated infrastructure options (private cloud routing)
White-label API access
Hardware-fleet management tools
Dedicated account management
Custom SLAs and support
How credits work:
Credits function like currency across platforms. Each platform sets their credit cost based on computational expense. The service takes a 20-30% margin on each transaction.
Infrastructure routing affects costs:
Mixed cloud (optimal): Base rate
Cheapest cloud: 15-20% discount (slower processing)
Greenest cloud: 10-15% premium (carbon offset included)
Local generation: 0 credits (you're using your own hardware)
Why this benefits everyone:
For creatives and studios:
Think of your current workflow as having accounts with 15 different utility companies—one for heating, one for cooling, one for lighting, one for appliances. Now imagine one universal account that powers everything.
✅ One subscription replaces 20+ - Single invoice, simplified accounting
✅ No credit waste - Unused Pika credits can be spent on Runway
✅ Infrastructure control - Choose speed vs. cost vs. sustainability
✅ Local + cloud hybrid - Maximize hardware investment, supplement with cloud
✅ Discovery without commitment - Try new tools using existing credits
For platform developers:
Currently, you're like a restaurant that also has to be a bank, a payment processor, AND a delivery service. What if you could just focus on being an amazing restaurant while someone else handled payments and logistics?
Platforms like Kinetix currently face:
€0.15-0.20 per emote generation
Most multi-discipline creatives won't subscribe for occasional use
Result: Limited addressable market - only power users convert
With universal credits:
✅ 10x user base growth The barrier drops from "another monthly subscription" to "already paid for credits." It's like going from requiring a Costco membership at every store to accepting the same membership card everywhere.
✅ Predictable revenue Payment guaranteed by credit provider. No chargebacks, no fraud management, no billing systems to maintain.
✅ Infrastructure flexibility Platforms can offer multiple performance tiers: "fast generation (100 credits) vs. economy generation (85 credits)." Users choose, you get paid either way.
✅ Discovery amplification Users try your tool because they already have credits. It's like test-driving cars when you already have a gas card versus having to set up a new fuel account for each brand.
✅ Resilience against the AI bubble When individual platforms rise and fall, you're plugged into infrastructure that persists. Like being a restaurant on a busy high street versus owning the entire mall—you want to be the high street.
For the service provider (the £10B opportunity):
This becomes infrastructure, not just a product:
Recurring revenue on billions of AI generations
20-30% margin on every transaction (like Stripe's 2.9% but for much higher transaction values)
Enterprise contracts with negotiated rates
Multiple revenue streams: credit subscriptions + infrastructure routing fees + API access
Why this hasn't been built yet
After extensive research, here's what exists and what doesn't:
What EXISTS:
✅ Chat/text AI aggregators - Magai, Poe, CosmicUp, Mammouth These solved language model access but missed specialized creative tools. Like building Netflix but only for documentaries.
✅ Walled garden credits - Adobe, Google, Microsoft These work within ecosystems but don't connect to third-party platforms. Like frequent flyer miles that only work with one airline.
✅ Enterprise cloud infrastructure - Oracle Multicloud Universal Credits This handles cloud compute but not creative generation credits. Like owning power plants but not retail electricity accounts.
❌ Universal creative generation credits - DOESN'T EXIST
❌ User-selectable cloud infrastructure routing - DOESN'T EXIST
❌ Local + cloud hybrid generation system - DOESN'T EXIST (ComfyUI has local-only, cloud platforms have cloud-only, nobody bridges both)
The gap is real. The opportunity is massive. And it's sitting in plain sight.
Who could actually build this?
Best positioned:
1. Envato (the frontrunner) They already executed Act 2 (Envato Elements unlimited subscriptions). They understand creative workflows. They have studio relationships. They just need to see Act 4 clearly.
It's like how Netflix went from DVD-by-mail to streaming—they had the customer relationships and understood the business model evolution.
2. Shutterstock / Getty (the incumbents) Massive B2B presence, enterprise legal frameworks, existing infrastructure. But ideologically resistant—they're still fighting AI instead of embracing it.
They're Blockbuster in 2009: everything needed to win, but unable to see the future.
3. Adobe (the walled garden) Perfect technical infrastructure but won't open up. They want you locked into Creative Cloud, not using credits across competitors.
They're Apple: incredible ecosystem, but refuses to play nice with others.
Dark horses:
4. Stripe (the infrastructure play) They revolutionized payment infrastructure by being invisible middleware everyone uses. Could do the same for AI credits. They understand margins, enterprise sales, and developer ecosystems.
Like AWS: nobody thinks about the infrastructure, but everyone relies on it.
5. Toolfarm (the dark horse) Established software reseller for creative tools with deep studio relationships. Could pivot from software sales to credit infrastructure. Small player, but perfectly positioned if they move fast.
Think Shopify: started selling snowboard equipment, became ecommerce infrastructure for millions.
6. A well-funded startup (the disruptor) Someone reading this with £10-50M in VC funding and a chip on their shoulder. No legacy business to protect. Pure infrastructure play from day one.
The Uber/Airbnb/Stripe opportunity: build infrastructure that seems obvious in hindsight.
Won't do it:
Individual AI platforms - Too focused on their own moat
Existing chat aggregators - They solved text AI, now they're stuck competing on that layer
ComfyUI community - Ideologically opposed to centralized systems (but someone could fork the concept)
What this requires to build
Let's be clear: this isn't a weekend hackathon project.
Technical infrastructure (12-18 months):
API integrations with 50+ platforms (custom for each)
Real-time credit tracking across distributed systems
Local generation detection and routing logic
Cloud provider partnerships (AWS, Google, Azure, etc.)
Hybrid processing orchestration
Fraud prevention at scale
Legal framework (6-12 months):
Tiered indemnification system (personal/commercial/enterprise)
IP licensing agreements with each platform
Multi-jurisdiction compliance (GDPR, CCPA, etc.)
Commercial use coverage and enforcement
Business development (ongoing):
Partnership deals with major platforms (20-30% margin negotiation)
Cloud infrastructure provider relationships
Enterprise sales team
Customer success infrastructure
Funding required:
£10-30M for credible MVP (single geography, 20 platforms)
£30-50M for global launch (multi-geography, 50+ platforms)
18-24 months to market with MVP
3-5 years to category dominance
But here's the thing: whoever does this first becomes infrastructure.
Not a product. Not a service. Infrastructure.
Like Stripe for payments. Like AWS for cloud compute. Like Shopify for ecommerce.
That's a £10B+ outcome.
Why I'm giving this away
I'm a Creative Director running a motion graphics and 3D design studio. I don't have £10M in the bank. I don't have enterprise sales relationships. I don't have a legal team to negotiate partnership agreements.
But someone reading this does.
This opportunity is too obvious, too valuable, and too beneficial to the entire creative industry to keep to myself.
Think about it:
Stock services went from fragmented to consolidated (Act 2)
Chat AI went from fragmented to aggregated (Magai, Poe, etc.)
Creative generation tools are fragmented RIGHT NOW
History doesn't repeat, but it rhymes. Act 4 is inevitable. Someone just needs to build it.
If you're a stock service executive, a VC looking for infrastructure plays, or an entrepreneur with the resources to execute—this is your moment.
The first mover advantage
We're at an inflection point similar to:
2010: When unlimited subscription models started (Envato Elements, Netflix)
2020: When chat aggregators launched (Poe, Magai)
2025: When creative generation needs universal infrastructure
The first company to crack this becomes the standard. Not because they're first, but because infrastructure has network effects.
More platforms = more value for users = more users = more platforms.
It's a flywheel. First mover builds the flywheel.
ComfyUI proved local generation in 2024. Chat aggregators proved unified access works. Cloud providers proved infrastructure routing is possible.
All the pieces exist. Someone just needs to assemble them.
What happens if nobody builds this?
The current fragmentation continues:
Creatives keep juggling 10-20 subscriptions (death by a thousand monthly charges)
Platform developers keep losing users to subscription fatigue
Innovation throttles because payment friction kills experimentation
Studios waste administrative overhead managing multiple vendors
Meanwhile, we keep throwing away money on unused credits while simultaneously running out of credits on platforms we actually use.
It's like having a fridge full of food while starving because nothing's in the right place.
Someone will fix this. The only question is who, and when.
Final thought
I've spent 15+ years in motion graphics and 3D design. I've worked with XPRESSO nodes in Cinema 4D, Redshift, ComfyUI, and countless AI platforms. I've seen technological shifts come and go.
This one's different.
The universal AI credit system isn't just a business opportunity. It's infrastructure that should exist.
The stock industry already walked this path:
Act 1 (1990-2010): Physical to digital (CDs to downloads)
Act 2 (2010-2020): Pay-per-item to unlimited subscriptions (Envato Elements)
Act 3 (2020-now): Mixed human + AI content (Adobe Stock hybrid)
Act 4 (Not yet built): Universal AI generation credits + infrastructure choice
We're living in the moment between Act 3 and Act 4. Someone just needs to build the bridge.
So here it is, named clearly:
Build the Stripe for AI credits.
Not just credits. Credits + cloud choice + local hybrid processing + infrastructure routing.
Be the electricity grid for AI generation.
Someone will. The only question is who gets there first.
Simon Spencer Harvey Creative Director, SSH Creative Ltd Newcastle upon Tyne, UK
If you're positioned to build this, or know someone who is, let's talk. If you're a platform developer who sees the value, let's talk. If you're a creative who's tired of subscription juggling, share this and make some noise.
The opportunity won't wait forever.
Comments