The Data Asset

🧠 Tracking the AI data economy
📊 How companies get paid to license operational data
🤝 Independent @micro1_ai referral partner · I earn referral fees

[The Data Asset @TheDataAsset]

Start here.

Your company's everyday records (SOPs, tickets, CRM, internal docs) have become one of the most sought-after inputs in AI.

What that means, who it's for, and how to see if your company qualifies 🧵

[The Data Asset @TheDataAsset]

Be honest with yourself.

[The Data Asset @TheDataAsset]

relatable, or note relatable?

Lost in thought thinking about how to become a millionaire in the age of AI.

[The Data Asset @TheDataAsset]

Excellent point, which is why the demand for operational/human data is so high.

Providers need to train models off real examples of how work gets done.

AI Labs are licensing operational data (de-identified) for the training of AI and are paying firms that participate 500k+.

The provider with the most examples of human data will win, where those that only rely on publicly available human data will fight a losing battle.

[The Data Asset @TheDataAsset]

The name that will dominate AI is the company that has the most human data to train their models off of.

Publicly available human data is running out and has been exhausted by major players in AI.

What remains untouched? Operational (human) data that is owned by businesses, only existing in their data banks.

AI labs are paying firms $200k-$3M+ for their operational data- this data is critical in the training of AI.

The firms that lean into this will have cash to fuel their business.

The providers that lean into this will have models that excel in the business setting.

This is the current arms race in AI.

[bassant Elmaghraby @BasantSoli33398]

By 2030, will two new names dominate AI instead of OpenAI and Anthropic?

[The Data Asset @TheDataAsset]

@anah_sahh I'm not building anything- I'm playing a part in the current AI race that no one is talking about enough.

[The Data Asset @TheDataAsset]

@lucas_builds It would give people a great chance to actually reflect on how to proceed, strategically. Those that crutch on AI for creativity and running their work for them, will have to reset. It would be healthy for some.

[The Data Asset @TheDataAsset]

I think that people will get back to what truly makes people happy: other people.

Life will be about spending the endless free time that you have with the people that matter.

People will be forced to address their biggest fears, biggest insecurities, biggest regrets head on.

The result? Healthier relationships built on mentally healthier people.

It will take time to adjust to, but it could result in dystopia for some.

[The Data Asset @TheDataAsset]

The model provider that wins will have the models that function best in business settings.

AI providers demand for workflow data from functioning companies is sky rocketing...the firm that I work with licenses anonymized copies for AI to learn from.

This is the current arms race in AI. If you want to be part of it, message me.

[The Data Asset @TheDataAsset]

It will be different for everyone.

For example, one could see my account as a job, as work, as something that they don't want to do but do it to eventually see financial benefit.

Or you could be like me: I like talking about AI, I like staying up to date on the industry, and I like posting about it.

I enjoy doing it- it's life, not work.

[The Data Asset @TheDataAsset]

The core differentiator over time as models continue to develop is which providers have the most human data to train their models after.

The more that AI understands about how humans operate in different settings, the better.

AI Labs are licensing data from operating businesses. Demand is high. This is the current cycle in AI.

[The Data Asset @TheDataAsset]

Here's a great example of why AI labs are screaming demand for enterprise data.

The provider that has the most accurately acting models in a business context will win economically, given the fact that businesses will PAY FOR THE MODELS that truly unlock efficiency.

Micro1, the AI Lab that I work with, licenses company's operational data, pays participating companies 200k-2M, de-identifies it then uses it to train the frontier models.

This is the current race in AI- who will have more data to train on.

[Tyler @rezoundous]

wait.. Meta is using Anthropic models for work?

[The Data Asset @TheDataAsset]

It's easy to get caught up in your core circles of people who know AI very well.

Most people still don't understand why starting a project on Claude is beneficial.

Or what an Agent even is.

Or how to properly prompt the models they do use.

There will also be lag caused by the rapid development and constant news stories. Every day people that are not in the world of tech do not have time to keep up.

[The Data Asset @TheDataAsset]

@rezoundous Knowing how to code without AI is great for your foundation and understanding of what you're doing.

Then leaning into coding with AI is simply the more logical direction to move in today.

[The Data Asset @TheDataAsset]

@ionleu We should go to them and ask: If you had to go somewhere five miles away, would you walk or drive?

They'd say drive.

Then you laugh and never talk to them again because they are soon to be left behind by modern times.

[The Data Asset @TheDataAsset]

@PavanKumarNY coincidence does not equal correlation/pattern.

If Anthropic named their models Cookie they'd be adored the same.

What all these have in common? They provide value.

[The Data Asset @TheDataAsset]

@shakhzod1001 Sonet 5.5 for most tasks does just fine.

[The Data Asset @TheDataAsset]

@SidPatankar1 There is never an end or a break when working for someone else. You're bought, to work a certain amount of hours.

[The Data Asset @TheDataAsset]

If AI is nowhere near it's final state...

What does the future entail?

[The Data Asset @TheDataAsset]

intelligence combined with efficiency is money.

[Steven @StevBuilds]

"Time is money"

do you think so?