Meta introduces new security features against fraud on Facebook, WhatsApp and Messenger

Almost everyone spends time in chat apps every day or scrolls through social networks. That is exactly where fraudsters strike, as they can most easily catch people out in the middle of daily life, between work, family and leisure time. Meta is now responding by equipping its platforms with new security features designed to identify suspicious contacts sooner and give users clear warnings.

Meta responds to wave of fraud attempts on social networks

Whether it is a supposed parcel courier, an alleged investor or a “new friend” with an attractive profile picture, scams on Facebook, Instagram, WhatsApp and Messenger have become far more sophisticated in recent years. Criminals deliberately use social media to build trust, obtain personal information or lure people into investments and fake prize draws.

The scam often begins with an innocent-looking message or a friend request from a profile that appears genuine at first glance. In many cases, however, it is a fake account using stolen photographs, imitating friends and presenting an invented personal history.

Meta is now introducing new protection mechanisms intended to check suspicious contacts automatically and alert users directly to potential traps.

To do this, the company is relying more heavily on artificial intelligence and additional warnings displayed within its apps. The aim is for users to become suspicious earlier, rather than only after money has been lost or an account has been taken over.

Facebook: AI filters dubious friend requests

On Facebook, a newly enhanced AI system is intended to assess friend requests before they arrive in your inbox. Among other things, it checks:

• whether there are noticeably few mutual friends;
• whether the stated location or country does not match the information on the profile;
• whether the account was only created recently;
• whether similar profiles have previously been linked to fraud attempts.

If a request appears suspicious, Facebook will display a prominent warning. This should prompt users to pause, examine the profile critically once again, and delete or block the request directly if necessary.

The warning message is intended to enable an “informed decision”: accept, ignore or block the profile with one click.

Older people in particular, as well as users who use Facebook only casually, can be quickly drawn in by profiles that appear friendly. These new notices are meant to protect precisely this group without shutting down the social network altogether.

Why this feature was long overdue

Fake profiles form the foundation of many fraud schemes: romance scams, fabricated pleas for help from alleged relatives, fake job offers and invented prize draws. If such accounts are filtered out sooner, many scams collapse at the very first stage.

The AI operates in the background, learns from reported cases and continually adjusts its assessments. It will never be flawless, but it raises the barrier for fraudsters considerably.

WhatsApp: new alert level for account access

With WhatsApp, the emphasis is more firmly on account protection. Increasingly, fraudsters are not merely entering chats but attempting to take over other people’s accounts directly, for example following a phishing email, a manipulated QR code or a fake security notice.

Meta is introducing an additional warning feature for this purpose: if someone tries to activate a WhatsApp account on a new device, users should be notified more clearly and at an earlier stage. If such a notification appears unexpectedly, caution is essential.

If a login warning appears when you have not set up a new device yourself, you should become suspicious immediately and reject the access request.

Especially after phishing attacks in which login details or SMS codes have been intercepted, this additional layer can make the crucial difference. The attack is uncovered before the offender gains full control over chats and contacts.

How to strengthen your own WhatsApp protection

Alongside Meta’s new feature, it is worth checking the settings. Anyone who enables “two-step verification” protects their account with an extra PIN. This makes it far harder for an attacker to cause harm, even with intercepted SMS codes.

Risk	Measure in WhatsApp
Account takeover	Enable two-step verification
Phishing link	Do not click links in messages without thinking
Unknown caller	Check the number and block it if appropriate

Messenger: “advanced scam detection” for suspicious chats

Meta is launching another security layer for Messenger: a feature that identifies scam patterns in ongoing conversations and actively warns users. Meta refers to advanced detection of fraud attempts based on schemes that are already known.

If an offender uses typical wording, for instance by promising quick profits, high returns with no risk or particularly lucrative side jobs, Messenger is intended to respond. Suspicious links, untrustworthy cryptocurrency “investments” and questionable job offers also fall within its scope.

If Messenger detects a possible scam, users receive background information on common schemes and specific suggestions: end the conversation, block the contact, report the account.

The system is primarily aimed at three particularly widespread tricks:

• Fake prize draws: Prizes or vouchers are promised but are only “unlocked” after personal details are entered or an advance payment is made.
• Fake investments: Unknown individuals aggressively promote financial investments, trading groups or supposed “insider tips” offering dream-like returns.
• Dubious job offers: Jobs promising “easy money”, often advertised on Instagram or other platforms, lead people into a trap.

The warnings appear directly in the chat window. Users can therefore see, in the context of the conversation, why the algorithm has become suspicious and then decide for themselves how to respond.

What users should do now

These new security features work only alongside attentive users. A number of basic rules remain essential, regardless of how much AI operates in the background.

• Never share personal data such as photographs of identity documents, TANs or passwords via chat.
• Scrutinise friend requests from complete strangers, even if the profile photograph and name look respectable.
• If money is requested, even by supposed friends or family members, insist on calling them back.
• Read unfamiliar warnings or security messages slowly, rather than hurriedly clicking buttons.
• Report suspicious messages and profiles directly instead of simply deleting them.

Fraudsters in particular rely on speed: pressure, supposed deadlines and “last chance” offers. Anyone who pauses for a moment and takes warnings from Facebook, WhatsApp or Messenger seriously removes an important advantage from criminals.

Why AI is both a blessing and a curse in the fight against fraud

Meta uses AI to detect fraud, but criminals have long been using AI too, to make their schemes seem more credible. Generated profile images, perfectly phrased messages and deceptively authentic fake websites are increasing the pressure on platform operators to keep refining their detection systems.

For users, this means that fraud attempts now often look much more professional than they did only a few years ago. A message in flawless German with a seemingly plausible story is no longer proof that it is genuine.

It is useful to know a few common terms. “Phishing” describes attempts to steal login details or confidential information through fake messages and websites. With forged QR codes, the codes are manipulated so that they lead not to the genuine website of a parcel delivery company or bank, but to a copy designed to capture data.

Meta’s new measures remove some of the burden from users by analysing suspicious patterns in the background. However, they are no substitute for reckless carelessness. Anyone who adopts a few simple security routines makes life extremely difficult for fraudsters, even when a scam appears perfect at first sight.