BLOG

Is poker RTA detectable? How sites catch it in 2026

GTO MAX AI team · 4 October 2026 · 7 min read

Every player who considers real-time assistance asks the same question first, and the honest answer is more interesting than yes or no. Sites catch RTA in five different ways, they are good at some and poor at others, and the method that catches the most people is the one no software can hide from. This is what is publicly known about each, written by a team that builds an RTA and wants its users to understand the ground they stand on.

1. Process and window scans

Desktop poker clients run with enough privilege to list the processes on your computer and the titles of open windows. Every major client does this. The lists they match against are not public, but security reports from GGPoker, PokerStars and Winamax over the years describe bans "for running prohibited software", which is this check. It catches installed tools: a solver running beside the table, a known RTA executable, screen-capture helpers. It does not see a browser tab, which is why the current generation of tools runs in a browser, and it does not see a second computer at all.

What people miss: this check is cheap and decisive, which makes it the first line, but it is the one that catches the fewest people, because anyone serious stopped running tools on the poker machine years ago.

2. Timing

A human's decision time varies with the difficulty of the spot. A player consulting a tool shows a different shape: a near-constant delay on every street, or a delay that tracks the tool's response time rather than the spot. Sites model timing per player and flag distributions that do not look human. The best RTAs vary when the advice appears and leave the clicking to the player, so the player's natural variance comes back; a tool that makes you wait for its answer every time is a timing signature.

3. The statistical fingerprint

This is the method that catches the most players, and no software can defeat it. Over thousands of hands, a player's frequencies converge on something. A human converges on a human strategy: a few leaks, a few spots played better than theory, a style. A player following solver output converges on the equilibrium, in every spot, with an accuracy no human reaches. Sites compare each player's frequencies to equilibrium benchmarks across hundreds of situations, and a player too close to all of them, for too long, is flagged for review. GGPoker's security team has said in interviews that this is their primary RTA signal.

Two consequences follow. First, the question "can they see my browser tab" is the wrong question: they are not looking at your computer, they are looking at your play. Second, a tool that deviates from the equilibrium on purpose, towards the exploits a human would also find, produces a less equilibrium-shaped record than a pure solver tool. That is not a reason to think any RTA is undetectable. It is a reason to understand that "plays pure GTO" is itself the fingerprint.

4. Device, network and account signals

Multiple accounts from one device, one account from many countries, virtual machines, remote-desktop sessions, and screen-sharing software that is active while a table is open are all logged. Screen sharing is the interesting one for browser RTAs: a browser tool works by sharing the table window, and some clients can see that a window-capture session exists even if they cannot see where it goes. Public information is thin here; assume the client knows the table is being captured.

5. Other players

Reports from opponents start more investigations than any automated signal. A player who never makes a mistake, who plays four tables with equal precision, who always sizes to the solver's number, gets reported, and a report triggers the statistical review above. Quiet play is not a technical property of a tool, it is behaviour.

6. What this means in practice

Where RTA is forbidden, every method above is live, and the statistical one has no counter. Accounts caught lose their funds. Treat the rules as real.
Where it is allowed, which is study, coaching, home games on sites like PokerNow, private clubs that permit it and play-money tables, none of this matters, and that is where a learning tool earns its keep: you see the equilibrium line, the exploit and the reason, hand after hand, until you no longer need them.
Browser-based tools remove the process-scan risk and nothing else. Anyone selling "undetectable" is selling the part of the problem that was already solved and hiding the part that cannot be.

How an RTA actually reads a table and solves a spot, and how the tools compare: the poker RTA guide. How GTO MAX AI deviates from the equilibrium per opponent: player profiling.

GTO MAX AI: the RTA that knows the player.

GTO line, exploit line and the reason, live, on up to four tables, in Chrome. GGPoker, Natural8, ClubGG, PokerNow, and PokerStars in beta. Try it for 30 minutes on Telegram.

Get a 30-minute trial

Coaching and study tool. Use it where real-time assistance is allowed: study, home games, private clubs, play money. Most online sites forbid real-time assistance. Advice can be wrong; every decision is yours. Terms of use.

More from the blog

Longer reads: the poker RTA guide and all guides.