Money · Operator memo

Does Article 50 apply to my AI email support agent?

Yes if the AI itself sends the email to the customer. The Commission FAQ says Article 50(1) needs a genuine two-way exchange and a direct interaction: the AI communicates with the person, not through a human intermediary.

A draft the AI writes that a human meaningfully reviews and sends is usually out of Art. 50(1). An agent that auto-replies in your brand name is in. Notice has been enforceable since 2 August 2026. A branded signature that looks like a person does not make it “obvious.”

Get the $29 Article 50 Ship Checklist Buy a Money Brief $149

Not legal advice. Related: ChatGPT wrappers · Exact notice words · Free notice tool

The four FAQ criteria, applied to email

The Article 50 FAQ (last update 24 July 2026) lists four cumulative criteria for the chatbot-style notice duty. All four have to hold:

  1. The system is an AI system.
  2. It is designed for a genuine two-way exchange with people, not merely collecting data or providing automated responses.
  3. The interaction is direct: the AI itself communicates with the person, not through a human intermediary.
  4. The person is a natural person (consumer, professional, or other user).

Email support agents that read a ticket, pull account data, and write a specific reply usually clear (1), (2), and (4). Criterion (3) is the fork: who actually hits send.

Who hits send

What you shipped Art. 50(1) notice? Why
AI reads inbox and auto-sends the reply under your brand Yes Direct AI-to-person communication. No human intermediary.
AI drafts; a human reviews the substance and sends Usually no under Art. 50(1) FAQ: interaction is not direct when it goes through a human intermediary.
One-shot “we got your email” autoresponder with no exchange Usually no FAQ carves out systems that only provide automated responses / collect data.
Classifier that routes tickets to humans, never writes to the customer No under Art. 50(1) Background / no direct contact with people.
Signed “Alex from Support” AI reply that looks human Yes, and treat “obvious” as out FAQ: interpret the obviousness exception restrictively.

Do not bury the disclosure in a privacy policy or Terms. The FAQ says people must be informed from the start of the first interaction, in a clear and distinguishable way.

What about Article 50(4) text labels

Separate duty. Article 50(4) is for deployers who publish AI-generated or manipulated text to inform the public on matters of public interest. Ordinary private support email is usually not “published” public-interest text. Support replies that never leave a private ticket thread are a different question from Art. 50(1) direct interaction.

If you later publish AI-written help-center or status posts on public-interest topics without real human review or editorial control, Art. 50(4) can kick in. The FAQ’s human-review bar is substance review, not spell-check.

What is live this week

Article 50 applies from 2 August 2026. The only grace period in the FAQ is for Article 50(2) machine-readable marks on generative systems already on the market before that date: those marks are due 2 December 2026. Chatbot / agent notice had no extra delay. The Digital Omnibus moved high-risk dates, not Article 50.

National market surveillance authorities enforce most Article 50 cases. Fines can reach €15 million or 3% of worldwide turnover, with proportionality language for SMEs.

Do this in seven days

  1. Inventory every surface where an AI writes to a customer: email, ticket replies, SMS, in-app messages.
  2. If the AI auto-sends, ship a clear first-interaction disclosure in that channel (subject line, first sentence, or persistent label). Commission-style example: “You are interacting with an AI system.”
  3. If a human always reviews and sends, document that intermediary step. Do not call a rubber-stamp “review.”
  4. Kill human-named signatures on auto-sent AI mail unless the notice is unmistakable.
  5. Open an Article 50(2) marking ticket only if you also generate synthetic audio, image, video, or text that people see outside private mail. Pre-2 August products get until 2 December for that mark.

Buy the checklist

The $29 Article 50 Ship Checklist is the print-and-tick page: inventory, notice line, deepfake labels, 2 December mark. The $149 Money Brief is the sourced memo if your stack mixes auto-send agents, human-in-the-loop drafts, and public help-center posts and legal wants the FAQ criteria applied line by line.

Get the $29 Article 50 Ship Checklist Buy a Money Brief $149

Free notice-copy tool · Omnibus did not delay Article 50

Sources

Fetched 28 August 2026 (PT). No invented quotes.

  1. European Commission, “Transparency obligations under Article 50 of the AI Act,” FAQ, last update 24 July 2026 (four Art. 50(1) criteria including human intermediary and “automated responses” carve-out; obviousness exception read restrictively; applies from 2 August 2026; 2 December grace only for Art. 50(2) marks; Art. 50(4) human review vs spell-check; fines up to €15m / 3%). digital-strategy.ec.europa.eu
  2. European Commission, “Commission starts enforcing AI Act rules and new transparency requirements on 2 August,” press release, 31 July 2026. digital-strategy.ec.europa.eu
  3. European Commission, AI Act policy page, last update 3 August 2026. digital-strategy.ec.europa.eu