Ofcom has outlined four ways social media platforms could check whether users are at least 16, including photo-ID matching and facial age estimation. The proposals put practical questions about access, privacy and the limits of automated age checks squarely on the table.
Watch Desk analysis
What happened
The UK regulator has narrowed its shortlist to four methods of highly effective age assurance as it prepares for the proposed under-16 social media ban, The Register reports. The options are photo-ID matching, bank checks, digital ID services and facial age estimation. Ofcom says facial estimates have a margin of error of several years and may need a second check.
Ofcom estimates that around 94 per cent of 16- and 17-year-olds have photo ID, while about 150,000 do not. It also says age-inference models could help platforms identify users who bypass checks. The ban is proposed, not in force, and the report says the government expects to set out its first regulations by the end of the year.
Why it matters
These checks are not interchangeable. Some rely on identity documents or banks, while facial estimation tries to infer age from a person’s appearance. Ofcom says that method is not reliable enough to stand alone. Its proposals also raise a practical access issue: young people without photo ID may have to obtain it to use services covered by the rules.
The report also records concerns from privacy campaigners about people having to prove their age to access online services. Ofcom acknowledges that circumvention cannot be eliminated entirely. The policy therefore faces two tests at once: whether checks can meaningfully restrict under-16 access, and what costs or privacy trade-offs they impose on everyone else.
Our read
Age estimation is a concrete use of automated inference in a consequential online policy, but it is not a magic bouncer. Ofcom itself says facial estimates can be several years out, and the report notes that children may still find ways around checks. The useful question is not simply whether platforms can ask for proof, but whether the method is accurate, proportionate and accessible to people who lack standard ID.
What to watch
- The government’s first regulations, expected by the end of 2026.
- Whether platforms combine facial age estimation with another check.
- How the rules address users without photo ID and attempts to bypass checks.
- What safeguards apply to identity data collected by platforms or verification services.
Discussion spark: Should platforms be allowed to use facial age estimation as part of these checks, or should age verification rely on methods that do not assess a person’s face?
Sources and evidence
Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.