A verification-first model for cloud Bitcoin mining

Cloud mining should be simple: an operator runs machines, rents out a share of their hashrate, and passes the mined coins to the renter. In practice, the category has been overrun by fraud, and the fraud follows a recognisable pattern.

• ·Simulated balances that climb on a fixed timer, imitating “earnings” that no real mining produces.
• ·Guaranteed daily returns — a percentage that mining, which depends on price and difficulty, can never promise.
• ·Withdrawals gated behind new deposits, the defining move of advance-fee fraud.
• ·Referral rewards paid on deposits, funding early payouts from later victims — the mechanics of a pyramid.
• ·Impossible coin lists, including “Ethereum mining” years after Ethereum stopped being mineable.

The damage is not limited to the scams themselves. Every honest operator inherits the distrust they create. Minevana’s response is not to argue that we are different — it is to build the platform so that a sceptical customer can prove it, using sources we do not control. The rest of this paper is that design.

Four principles govern every decision on the platform. Where a growth tactic conflicts with one of them, the tactic loses.

The platform is a pipeline from physical machines to a customer’s wallet. Each stage is designed to expose a checkable fact rather than an opaque promise.

“AI mining” is often marketing noise. Underneath it, machine learning does real, measurable work in exactly four places — all of them about cost and reliability, never about conjuring yield.

• ·Dynamic tuning. Per-chip voltage and frequency curves that minimise joules per terahash for each machine and its ambient temperature.
• ·Predictive maintenance. Telemetry models flag failing fans, hashboards, and power supplies before they cause downtime.
• ·Energy optimisation. Shifting flexible load to cheaper power windows and participating in demand-response where available.
• ·Profit routing. For multi-algorithm hardware, directing hashpower to the most profitable coin or pool, net of fees.

The efficiency metric is joules per terahash (J/TH) — lower is better — and small, persistent gains compound into a materially lower cost base over months. The critical trust point: the model optimises a physical, measurable outcome (energy per unit of work, error rate, uptime), not a number on a customer’s screen. What AI cannot do is equally important — it cannot exceed the hardware’s physical hash rate, and it cannot control Bitcoin’s price or network difficulty. It lowers cost; it does not remove market risk.

This is the core of the design. For each claim Minevana makes, there is a way for the customer to confirm it independently — on a source outside our control.

The reason this works is the non-custodial architecture. Because coins are paid to your wallet rather than credited to an internal balance, there is nothing for an operator to freeze, fake, or lose. The most common failure mode of cloud mining — a balance that exists only inside the app — is structurally impossible here.

A Minevana plan is a purchase of a fixed amount of hashrate for a term — a computing service, not a deposit, a savings account, or an investment contract. What you buy is stated in TH/s, the same unit professional miners use.

Founder pricing, shown in full on the plans and checkout page before any purchase.

The fee formula

Pricing is transparent by construction: plan price, plus an electricity-and-maintenance fee quoted up front, plus the pool fee passed through at cost. There is no hidden spread on payouts and no fee invented at withdrawal time.

Payout mechanics

Rewards derive from the actual output of the pool your hashrate contributes to. They are paid to your wallet, each with an on-chain hash. The figure shown as “earned” on your dashboard is the sum of real payouts you have received — it never accrues on a schedule and never moves without a matching transaction. Plans can be bought with BTC, ETH, BNB, SOL, and USDT/USDC across major networks; the receiving address is shown per coin and network at checkout.

Minevana runs a two-tier affiliate program: 5% on tier 1 (people you refer directly) and 3% on tier 2 (their referrals). Three design choices keep it a legitimate affiliate program rather than a pyramid:

• ·Commissions are calculated on plan purchases — never on deposits and never on anyone’s mining income.
• ·They are funded from our sales margin, not from other customers’ money.
• ·There are exactly two tiers, and commissions clear after a 14-day window before they become withdrawable (and reverse if the underlying purchase is refunded).

An affiliate program pays for a real product sale out of the seller’s margin. A pyramid pays earlier participants out of later participants’ deposits. The distinction is not cosmetic, and Minevana is built firmly on the first side of it.

The platform’s security rests on standard cryptographic practice and, above all, on holding as little of value as possible.

• ·Password storage. Passwords are hashed with PBKDF2-SHA256 (100,000 iterations) using a unique per-user salt. Plaintext passwords are never stored.
• ·Sessions. Session tokens are HMAC-SHA256 signed and carried in HttpOnly, SameSite=Lax, Secure cookies. The signing secret is fail-closed: if it is absent, no session can be issued or validated, so a misconfiguration becomes a safe outage rather than a forgeable-session compromise.
• ·Rate limiting. A sliding-window limiter throttles abuse. On login it counts only failed attempts and clears on success, so a correct password is never locked out.
• ·Transport and headers. HSTS, a strict Content-Security-Policy, X-Frame-Options: DENY, X-Content-Type-Options: nosniff, a locked-down Permissions-Policy, and a strict Referrer-Policy are enforced site-wide.
• ·Architectural minimisation. Because payouts are non-custodial, there is no pooled customer balance to steal — the single most valuable target in most crypto platforms simply does not exist here.

Mining carries real financial risk, and we would rather lose a sale than a customer who did not understand it. In summary:

• ·Variable income. Rewards depend on network difficulty, block subsidy, fees, luck, and uptime — none of them fixed.
• ·You can receive less than you paid. If price falls or difficulty rises during your term, mined value can be below the plan cost.
• ·Price volatility. Payouts are in cryptocurrency, whose fiat value can drop sharply, including after payout.
• ·Operational and counterparty risk. Hardware fails, power prices move, and any cloud service means trusting an operator — which is precisely why the design leans on independent verification.
• ·Regulatory risk. Rules differ by jurisdiction and change; you are responsible for the tax treatment of mined coins.

Nothing on Minevana is investment, legal, or tax advice; we sell computing services. The full statement is in the risk disclosure. Do not spend money you cannot afford to lose.

The clearest way to describe an honest platform is to name the dishonest practices it rejects. These are commitments, not preferences.

Priorities are ordered by what earns trust, not by what looks impressive.

Minevana is operated by a team focused on transparent, efficient Bitcoin mining. Our registered legal entity, leadership, facility footage, and third-party attestations are published on the site as each becomes verifiable — we would rather show a placeholder than a claim we cannot back. That same standard is why this paper commits, in writing, to the verification and “never do” sections above: they are the terms we intend to be held to. Questions are answered plainly, including about the risks, at hello@minevana.com.

ASIC

Application-Specific Integrated Circuit — a chip built to do one job. Bitcoin mining ASICs compute SHA-256 hashes and nothing else, which is why they are far more efficient than general-purpose hardware.

Hashrate (TH/s)

The rate at which mining hardware performs hashes, measured in terahashes per second. It is the unit you rent: more TH/s means a larger share of block rewards, on average.

J/TH

Joules per terahash — how much electricity a machine burns per unit of work. Lower is better; it is the number the AI layer is built to reduce.

Proof-of-work

The consensus mechanism Bitcoin uses, where machines compete to solve a computational puzzle to add the next block. Mining exists because of proof-of-work.

Block subsidy / halving

The new bitcoin awarded for each block — currently 3.125 BTC after the April 2024 halving, and roughly halving again around 2028. Lower subsidy makes efficiency decisive.

Mining pool

A group of miners that combine hashrate and share rewards proportionally, smoothing out the luck of solo mining. Public pool accounts let claimed hashrate be verified.

Non-custodial

A design in which the operator never holds your funds. Coins are paid directly to a wallet you control, so there is no internal balance to freeze or lose.

Block explorer

A public website that lets anyone view any transaction on a blockchain. It is how you independently confirm that a payout really happened.

• Risk disclosure — The full, plain-language statement of what can go wrong.
• About Minevana — Who we are and the principles behind the platform.
• Cloud mining with AI — How the platform works, end to end.
• AI crypto mining, explained — What AI genuinely does in mining — and what it cannot.
• AI Bitcoin mining — SHA-256, the halving, and efficiency in depth.
• AI mining calculator — Model results under your own assumptions before buying.