Can the DWP Really Check Your Social Media?
If you claim benefits in the UK, you have probably seen headlines warning that the DWP is "watching your Facebook" or "monitoring your TikTok." Some of that is based on fact, but much of it is exaggerated or misunderstood. This guide explains exactly what the Department for Work and Pensions can and cannot do with your social media, what the law says, and where the real risks lie for claimants in 2026.
The short answer is: yes, the DWP can look at your publicly available social media posts, and it does. But the way it works in practice is more nuanced than most headlines suggest.
What the DWP Actually Does with Social Media
The DWP uses what it calls Open Source Intelligence (OSINT) — publicly available information from social media platforms, websites, forums, and other online sources. Its internal guidance describes these as "an important source of information in combatting benefit fraud."
In practice, this means DWP fraud investigators can:
- View your public Facebook, Instagram, TikTok, X (Twitter), and other social media profiles
- Take screenshots of photos, location check-ins, and posts
- Use those screenshots as evidence in an interview under caution
- Build a picture of your lifestyle and daily activities based on what you post
This is not mass surveillance of every claimant. It typically happens when a fraud investigation has already been opened — often triggered by a tip-off, a data mismatch, or a referral from another part of the DWP.
The AI Contract: Capita Cosain
In December 2023, the DWP signed a contract with Capita for the use of Cosain, an AI-driven social media listening and OSINT platform. The contract was valued at £77,760 for three annual licences. According to Capita's product documentation, the software was developed with OSINT experts from law enforcement backgrounds.
The key facts about Cosain:
- It can automatically scan and flag social media posts at scale
- Capita's guidance indicates up to 2.2 million posts could potentially be flagged to the DWP annually
- Flagged data is automatically deleted after 30 days
- This was the first time the DWP deployed automated tech that could monitor social networks at large scale
This means the DWP is not simply relying on individual investigators manually browsing Facebook. It now has AI tools capable of scanning public posts across platforms, although the exact scope and current status of the programme is not fully transparent.
What the Law Says: RIPA and Surveillance Rules
The DWP's surveillance powers are governed by the Regulation of Investigatory Powers Act 2000 (RIPA). Under RIPA, the DWP is authorised to conduct directed surveillance only. It is explicitly not authorised to:
- Conduct intrusive surveillance (such as bugging your home or tapping your phone)
- Access your private communications data (such as private messages, DMs, or encrypted chats)
- Use Covert Human Intelligence Sources (CHIS) — essentially informants or people in your life acting as sources
There is an important distinction around repeat viewing. DWP guidance states that if investigators need to view your social media more than twice over a period of time to build up a picture of your activities, they should consider applying for a formal Directed Surveillance Authorisation under RIPA. A single viewing of a public profile generally does not count as directed surveillance, but repeated monitoring may.
What Counts as "Public"?
If your social media profile is set to public, anyone can see it — including DWP investigators. This includes your posts, photos, check-ins, and comments on public pages. If your profile is set to private, investigators cannot legally access it without a warrant or other legal authority. They cannot, for example, create a fake profile to send you a friend request — that would likely constitute a CHIS operation, which the DWP is not authorised to conduct.
What Investigators Actually Look For
When a fraud investigation is underway and investigators check social media, they are typically looking for inconsistencies between what you have told the DWP and what your online presence suggests. Common examples include:
- Lifestyle inconsistencies: Posting about expensive holidays, new cars, or luxury purchases while claiming means-tested benefits where you have declared low income or savings
- Work activity: Evidence of paid work (such as advertising a business, promoting freelance services, or posting about shifts) while claiming as unemployed or declaring no earnings
- Relationship status: Evidence of a partner living with you (such as tagged photos, shared location check-ins, or joint social activities) when you are claiming as a single person
- Physical capability: In PIP or ESA cases, photos or videos showing physical activities that appear inconsistent with the limitations described in your claim
A Word of Caution on PIP and Disability Claims
This is an area where context is critical. There are documented cases where DWP investigators have used screenshots of claimants smiling at a wedding or on holiday as evidence that they are not in pain. Disability charities and legal experts have repeatedly criticised this approach, pointing out that a single photograph does not reflect the reality of living with a fluctuating condition. A person with chronic pain can smile in a photo and still be in significant pain. A person with a mental health condition can post a positive moment online while struggling every other day of the week.
If you are a PIP or ESA claimant, you are not doing anything wrong by living your life and posting about it. But it is worth being aware that photos and posts can be taken out of context during an investigation.
The New 2025 Act: What Has Changed
The Public Authorities (Fraud, Error and Recovery) Act 2025 significantly expanded the DWP's investigative powers. However, the headline changes are about bank data, not social media. The key new powers taking effect from 2026 include:
- Eligibility Verification Measure (EVM): Banks will be required to flag accounts where savings exceed the £16,000 capital limit for Universal Credit, or where patterns suggest extended overseas stays. The DWP will not see your transactions or spending — only that a flag has been raised.
- Direct Deduction Orders: If an overpayment debt is established and you refuse to engage with a repayment plan, the DWP can instruct your bank or employer to deduct payments directly — without going to court.
- Search and seizure warrants: Investigators can now apply for warrants to enter premises and seize evidence, a power they previously did not have.
- Driving licence sanctions: Repeated failure to repay debts of £1,000 or more can result in a two-year driving ban.
These new powers are significant, but they apply to financial data — not to social media content. Social media monitoring continues under the existing RIPA framework and OSINT practices described above.
What Is Exaggerated or Misleading
There is a lot of fear-driven content online about DWP surveillance. Here is what is not accurate:
- "The DWP monitors every claimant's social media." Not true. Social media checks happen during investigations, not as routine monitoring of all 6+ million Universal Credit claimants.
- "They can read your private messages." Not true. The DWP is not authorised to access private communications under RIPA. Only publicly visible content can be viewed without further legal authority.
- "Posting a holiday photo means you will lose your benefits." Not true. Going on holiday does not automatically disqualify you from benefits. However, if you are claiming based on financial hardship and posting about frequent luxury holidays, that could trigger questions.
- "They use fake profiles to friend you." The DWP is not authorised to use CHIS (Covert Human Intelligence Sources), and creating a fake profile to befriend a suspect would likely fall into that category.
- "An algorithm can stop your benefits." The 2025 Act explicitly requires a human case officer to review any automated flag before action is taken. No algorithm can stop your payments on its own.
What You Should Actually Be Aware Of
None of this means you should delete your social media or stop posting. But if you claim benefits, it is sensible to:
- Review your privacy settings. Make sure you understand who can see your posts. On Facebook, Instagram, and TikTok, you can restrict your profile visibility to friends or followers only.
- Be accurate in your claims. The biggest risk is not social media itself — it is having a gap between what you have told the DWP and what your online presence shows. If your claim is accurate and honest, your social media is unlikely to cause problems.
- Understand what counts as a change of circumstances. If you start a new relationship, begin working, move to a new address, or your health improves, you are legally required to report this to the DWP. Failing to do so is the most common cause of overpayments and fraud investigations — not social media posts.
- Know your rights in an interview. If you are called for an interview under caution, you have the right to have a solicitor present. You do not have to answer questions, and anything you say can be used in a prosecution. Free legal advice is available through Citizens Advice or a duty solicitor.
The Bigger Picture: Fraud and Error in Numbers
It is worth putting this in context. In 2024/25, the DWP estimated that £9.5 billion was overpaid due to fraud and error — around 3.3% of total benefit spending. However, this figure includes both deliberate fraud and genuine mistakes by claimants or the DWP itself. Claimant error (not fraud) accounts for a significant portion of overpayments.
The government has set a target to save £1.5 billion by 2029/30 through the new Act, as certified by the Office for Budget Responsibility. Around 6,000 staff were carrying out fraud and error reviews by March 2025, with 1.15 million claims reviewed.
Civil liberties groups including Big Brother Watch and Privacy International have raised concerns that the expanding surveillance powers create a "two-tier system" where benefit claimants are subject to privacy invasions not applied to the rest of the population. The Information Commissioner has also questioned whether the bank-scanning powers are "proportionate."
Key Takeaways
- The DWP can view your public social media posts, and has AI tools to do so at scale
- This typically happens during a formal fraud investigation, not as blanket monitoring
- They cannot access your private messages, use fake profiles, or conduct intrusive surveillance
- The new 2025 Act expands powers around bank data, not social media specifically
- A human must review any automated flag — no algorithm can stop your benefits
- The biggest risk is inaccurate claims, not social media activity
- Review your privacy settings and report changes of circumstances promptly
Check What You Are Entitled To
Many people who worry about DWP investigations are actually underclaiming, not overclaiming. If you are unsure whether you are getting everything you are entitled to, use our free benefits calculator to check in under five minutes. You may find you are missing out on support you are legally entitled to — and claiming correctly is the best protection against any investigation.