Question 1

What is AI freedom?

Accepted Answer

AI freedom is the ability to use AI without depending on a single company that can take it away, read it, or change the rules on you. Concretely, it means open-weight models you can run anywhere, infrastructure with no single operator, privacy you can verify instead of trust, open-source code you can audit, payment that does not require handing over your identity, and governance that is not set by one vendor's roadmap. LibertAI assembles these into one stack so that no single party holds the off switch.

Question 2

Why does it matter who runs the AI model?

Accepted Answer

Whoever runs the model controls what it will discuss, what happens to your prompts, when the model is retired, and what it costs once you depend on it. When one company holds all of that, your access exists at their discretion, and their values, policies, and business decisions are applied to your work. Running on open models and decentralized infrastructure spreads that control out, so no single operator can read your data, ban your account, or deprecate the model you rely on.

Question 3

What happens when a provider deprecates a model you depend on?

Accepted Answer

On a closed platform, a deprecation is final: the weights live only on the provider's servers, so when they retire a model, the prompts and behavior you tuned around it stop working and you are pushed onto a replacement that responds differently. With open-weight models, deprecation is not the same threat. The weights are public, so a model that works for you can keep running on LibertAI or on your own hardware regardless of any one company's roadmap. You decide when to move on, not the vendor.

Question 4

How does decentralization protect AI access?

Accepted Answer

Centralized AI has a single point of control: one company owns the servers, holds the logs, and can suspend an account or change terms unilaterally. LibertAI runs inference across Aleph Cloud's decentralized network, where there is no single operator to ban you, no central log store to breach or subpoena, and no one switch to flip. Removing the single point of control removes the single point of failure, which is what keeps your access from depending on one company's permission.

Question 5

What is user-owned AI?

Accepted Answer

User-owned AI means the parts that matter belong to you rather than to a vendor: the models are open-weight and run anywhere, the privacy is enforced by hardware and verifiable through attestation, the code is open source and auditable, and the direction is set by community governance through the LTAI token. The practical test is whether you have an exit. With user-owned AI you can leave with your workflows intact because the models are not trapped on one company's servers.