Use cases

Seven concrete scenarios.

Not theoretical cases. Situations lived today by elected officials, journalists, citizens, companies, platforms.

Public figure

A mayor facing a deepfake

A local elected official in mid-campaign. An AI-generated video shows him uttering racist remarks. The video has been circulating on WhatsApp and Facebook for 6 hours. The official denial takes 18 hours to reach the same audience as the fake.

Before Alethea

·Official denial published on the city hall accounts.
·Picked up by local press, capped at its audience.
·Streisand effect in reverse: those who saw the fake do not see the denial.
·Doubt seeded in undecided voters.

With Alethea

·Drop of the falsified video into the elected official's Alethea wallet.
·Select Reclaim role, add the context ("AI-generated video, here is the real statement signed on this date").
·Signature in under a minute.
·Any derivative version (recompressed, recut, reshared) triggers the rebuttal display automatically via compatible browser plugins.
·Knowingly distributing the fake becomes a traced act, potentially actionable (e.g. article 226-8 of the French Penal Code, or its international equivalents).

Journalism

A journalist verifies a viral video

A video allegedly filmed in Kyiv circulates on social media. Before relaying it, a journalist needs to know if it is authentic. Today: Google reverse image search, visual geolocation, peer verification. Hours of OSINT work.

Before Alethea

·Reverse image search, manual research.
·If the video is recent or original, no prior trace.
·Editorial decision taken under uncertainty.

With Alethea

·Upload the video into the Alethea verifier.
·Match in seconds against the public registry.
·Display: "signed by [author, eIDAS high], Witness role, March 12, 2026 at 14:32 UTC, in Kyiv, context provided".
·Editorial decision grounded, traceable, defensible.

Citizen

A citizen faces a dubious Facebook share

A friend shares a video of a political figure making shocking statements. The citizen wants to know whether it is real before sharing it further. Today they have no easy tool available.

Before Alethea

·No reliable consumer tool.
·Sharing decision based on gut feeling.
·Disinformation amplified by doubt.

With Alethea

·Alethea browser extension installed in one click.
·The extension automatically displays the signature status next to the video: signed, refuted, unsigned.
·If the content has been refuted by the person staged in it, the rebuttal is overlaid.
·The citizen shares knowingly, or chooses not to share.

Brand, organization

The Arup case, 2024: $25M stolen via deepfake

In 2024, the engineering firm Arup lost $25M during a video call where deepfake audio and video impersonated the CFO and several executives. A finance employee made 15 transfers in good faith.

Source: CNN / Hong Kong Police, May 2024

Before Alethea

·No live cryptographic verification mechanism.
·The visual and voice were enough to establish trust.
·$25M lost in hours.

With Alethea

·Alethea plugin integrated into the corporate video conferencing client.
·Live display: "participant signed (NIST IAL3, Sumsub Corporate), valid signature".
·Any unsigned participant is visually flagged and blocked from sensitive operations per the organization's policy.
·The deepfake no longer has access to the financial decision.

Platform

Native integration by a social platform

A video-sharing platform wants to offer its users a reliable trust signal, without breaking the open contribution model. The challenge: integrate Alethea without making signing mandatory.

Before Alethea

·A posteriori moderation via flagging.
·Unreliable AI detectors, false positives and negatives.
·No positive trust signal for authentic content.

With Alethea

·Alethea API called on upload for signed content.
·Badge "Alethea-signed — [standard, role]" displayed next to the content and to the author.
·For refuted content, automatic info banner with link to the rebuttal.
·Unsigned content remains shareable (visible fiction presumption).
·Progressive compliance with future legislative frameworks on signing of public content.

Personal trust

The WhatsApp from 'your boss' asking for an urgent wire

A WhatsApp voice message from your boss: indistinguishable voice, urgent tone, €15,000 to wire right now. Or a video call from your mother in distress asking for money. These deepfake-based fraud attempts grew +1700% between 2023 and 2025. The audio is cheap to produce (a 3-second voice sample is enough), the video is becoming so. The current defense ("did this sound a bit off?") is breaking down.

Before Alethea

·No technical way to verify the sender's voice or face is real.
·You act on instinct, often under social pressure ("urgent", "I can't talk now").
·The bank notices the issue after the wire is gone.
·The real sender (boss, parent) only finds out hours later.

With Alethea

·Your contacts (boss, parent, CFO) use an Alethea wallet at the identity standard level they choose (eIDAS substantial for individuals, high for executives handling money flows).
·Important voice messages, video calls, and authorizations are signed at production time, role 'author' (or 'actor' if the sender appears in the video).
·Your messaging app has an Alethea-compatible verifier (plugin or native integration). On reception, the message shows: "Signed Alethea — [Boss Name], eIDAS substantial, 14:32 UTC".
·If unsigned: a clear "Unverified content, presumption of fiction" badge.
·Critical: if your contact normally signs and this specific message does not, a red flag appears: "This contact usually signs their messages. This one does not. Be cautious."
·For high-value actions (wire transfers above a threshold, password resets), the verifier can refuse to display the request if the signature is missing or weak.

Bank, finance

The bank that refuses to wire €50,000 without a signed instruction

Push-payment fraud (where the victim authorizes the wire after being manipulated by a deepfake or phishing call) cost European banks over €8 billion in 2024. Once the wire is gone, recovery is rare. Banks need to block these flows before they leave, but the customer is the one giving the order, so traditional fraud detection has nothing to act on. Alethea provides the missing layer: cryptographic proof that the customer is the actual author of the order, signed in their wallet, not under coercion.

Before Alethea

·Wire orders are validated via SMS one-time code or app push, both phishable.
·If a customer is manipulated and types the OTP themselves, the bank sees a perfectly valid order.
·Vishing (voice phishing impersonating bank staff or relatives) bypasses every voice-based callback procedure.
·Push-payment fraud disputes are typically rejected by the bank because the customer technically authorized the transaction.

With Alethea

·For wire orders above a configurable threshold (e.g. €5,000, or any cross-border transfer), the bank requires an Alethea signature from the customer at identity level eIDAS high (FranceConnect+, Itsme, eID-AS).
·The signature is produced in the customer's wallet, with the order details (amount, beneficiary IBAN, date) cryptographically bound via the contextHash. Tampered details break the signature.
·If the signature is missing or below threshold: order rejected, customer asked to re-sign in their wallet.
·Bank-side audit: every high-value wire carries an on-chain proof that the customer signed exactly this content, at this time, with their verified identity. Trivial to defend in fraud disputes.
·For inbound calls claiming to be the bank: the bank itself signs its critical client communications (statements, alert SMS templates, key emails) with role 'broadcaster'. The customer's banking app verifies and displays a badge: real bank, or impostor.
·Compliance bonus: meets the upcoming PSD3 / EU AMLR strong customer authentication requirements at a higher bar than current SCA, with cryptographic non-repudiation.

Understand the protocol Become a pilot