Telegram widgets under scrutiny: Kaspersky warns of suspected WebRAT delivery

Kaspersky has warned that attackers may have used Telegram's newly launched widgets to deliver WebRAT malware. The warning followed public remote code execution (RCE) demonstrations by the security firm V12. Telegram CEO Pavel Durov has rejected the claims, saying no one could be hacked just by receiving a message and that the issue was fixed on Telegram's servers.

Security News 4 min read

Telegram widgets under scrutiny: Kaspersky warns of suspected WebRAT delivery, while Telegram disputes the remote code execution claims

On 9 November 2026, Telegram introduced HTML widgets, letting bots send small interactive apps in chat messages. Telegram's Bot API 10.4 release enabled this.

On 11 November, researchers at V12 published a video showing a widget message opening the Calculator app on an Android phone. They described this as remote code execution. A second V12 video, published later, claimed a similar exploit worked against Telegram's Windows client after the company's patch.

On 13 November, Kaspersky warned that visits to certain Telegram channels had triggered a PowerShell script. The script downloaded and launched WebRAT, a Rust-based trojan also known as Salat Stealer. WebRAT gives attackers control of an infected system and can target cryptocurrency wallets and account credentials. Kaspersky said the link to the widgets was only a possibility. It did not share technical details or say how many users were targeted.

The same day, Durov posted on X: "The issue doesn't exist. No one could be hacked just by receiving a message. You had to start a 3rd party service first. Even that is fully patched server-side." He also challenged researchers to try to hack his account.

Are you affected?

You may be at risk if you use Telegram widgets on Android, Windows or other clients, particularly if you open widgets from bot chats or channels you do not fully trust.

Telegram says the reported attack path required a third-party service to be started first, and that the server-side fix is complete. Researchers dispute how narrow that path is. Until the position is clear, treat widgets as an unverified feature.

What to do

  1. Do not launch widgets until Telegram or a trusted security vendor confirms that the issue is fixed on your platform. This is Kaspersky's main recommendation.

  2. Update Telegram on every device, including desktop and mobile. Secondary reports say client fixes were released in November 2026, and the latest version should include them. Check the version notes in your app store.

  3. Watch for unexplained PowerShell activity on Windows machines that use Telegram. A PowerShell process that starts after opening a Telegram channel or bot is worth investigating.

  4. Check for WebRAT indicators, such as unexpected processes, new scheduled tasks or unknown software, and look at what the trojan could reach, including cryptocurrency wallet files and saved passwords.

  5. If you suspect infection, disconnect the machine from the network, change passwords for important accounts from a clean device, and enable two-factor authentication on Telegram and any linked accounts.

  6. Keep the widget feature disabled if your organisation has a policy on messaging apps, and tell staff not to open widgets from unknown bots.

Why the dispute matters

The argument is about scope rather than whether the flaw existed. V12 described the issue as remote code execution through messages. Durov argues that the attack needed a third-party service to be active first, which sharply limits who could be affected.

Both positions can be partly true. The first is that the widget feature gave a new route into the app. The second is that the route depended on user actions, such as opening a bot chat. Neither is a reason to ignore the warning, because even a narrow attack path can be used at scale against users who open unknown bots.

There are also unresolved questions. Telegram has said it patched a related issue in version 12.10.4 in late September, before the research was published. Reports on the server-side changes differ, and some outlets have reported that Telegram disabled HTML widgets on its servers on 13 November. Telegram has not confirmed that in an official statement.

The lesson

New features expand what a messaging app can do, and that expansion creates new attack surface. Interactive content inside messages should be treated as code that can run on your device, not as simple text. Until a feature has been tested and stable for some time, most users and organisations should wait before enabling it.

Blog

Security 4 min read

N-central under attack: 4EverAble2Shell exploit chain is installing crypto miners on unpatched and patched servers

Attackers are exploiting 4EverAble2Shell, an unauthenticated exploit chain against N-able N-central, to gain root access on on-premises servers and install cryptocurrency miners. Researchers reported at least 21 compromised servers within hours of a public proof-of-concept appearing. Nineteen of them were running the latest hotfix, so the fix alone is not a complete answer. If your organisation runs on-premises N-central with the Legacy API enabled, treat any exposed server as a priority.

Microsoft 4 min read

Fake "Outage" emails follow Microsoft 365 disruption: check for held-message lures before you click

On 12 November 2026, Microsoft 365 and Outlook suffered a widespread outage. Downdetector recorded user reports from 74 countries and territories. Microsoft says the disruption was not caused by a cyberattack. Within a day, attackers began sending phishing emails that falsely claim messages were held during the outage and push recipients to a "restore messages" link. Microsoft has warned that these emails resemble credential phishing, but has not confirmed that any credentials were stolen.

AI 5 min read

149,000 Uber drivers exposed through a third-party AI provider: how a single set of credentials led to a major identity risk

Uber is notifying 149,346 drivers that an unauthorised party accessed their personal information through Fireworks AI, a cloud platform that runs and fine-tunes open-source AI models. The attacker used compromised credentials belonging to the provider to reach data Uber had shared with it. The exposed data includes names, addresses, dates of birth, driver's licence numbers, other government ID numbers and images of driver's licences.