A new briefing paper by free market think tank the Institute of Economic Affairs (IEA) says the government’s flagship online safety bill will have a significant impact on free speech, privacy and innovation.

The paper highlights the broad scope, complexity and reach of the bill, at over 255 pages, an increase of 110 pages since the May 2021 draft.

The bill’s proposed establishment of safety duties, under the threat of multi-billion pound fines and criminal sanctions, risks digital platforms using automated tools in a cautious and censorious manner. This will embolden bad faith actors to seek content removals.

The IEA also say the bill’s “byzantine” requirements will impose large regulatory costs, particularly on start-ups and smaller companies, and make it riskier to host content and develop innovative services. This could discourage investment, and cement the market position of “Big Tech” companies who can afford to comply.

The impact assessment estimates that the bill will cost businesses £2.5 billion over the first ten years. However this underestimates the direct costs by claiming legal fees are just £39.23 per hour, and does not attempt to assess the potential costs to innovation, competition or international trade.

There is a lack of evidence to justify the legislation, with respect to both the alleged prevalence of what the bill treats as “harm” and the link between the proposed measures and the desired objectives.

The paper claims that the bill’s new duties will impose unprecedented censorship on lawful speech, through removal of suspected illegal content to ensure compliance and duties in relation to content that is allegedly harmful but not illegal.

Platforms like Facebook, YouTube and Twitter will be expected to take action against speech if they “reasonably consider” it could be illegal, a significantly lower standard then “beyond reasonable doubt”.

If they fail to do so, the companies can be fined up to 10 per cent of global revenue. The new harms-based communication offence in the bill outlaws intentionally causing psychological distress. The IEA say this risks empowering the easily offended and bad faith actors to solicit removals of legal speech from platforms.

The Secretary of State will be able to direct Ofcom to change codes of practice “for reasons of public policy” and will set Ofcom’s priorities. Ofcom, through codes of conduct, will be able to decide what content could be “harmful” and thus set, and therefore potentially limit, the bounds of online free speech.

The report argues that the bill’s free speech protections are very weak – the bill only establishes a duty on platforms to “have regard” to the importance of protecting users’ freedom of expression.

Shadow DCMS Secretary Lucy Powell has even envisaged an extension of the notion of “harmful” to cover matters of public policy debate, having raised concerns that the bill would allow “climate deniers” to “slip through the net”.

The bill’s safety duties require general monitoring of user content using automated tools, which the report says amounts to mandate on platforms to actively assess every element of user speech, including private messaging apps such as WhatsApp and Signal.

The requirement for platforms to undertake age assurance and verification could mean users regularly being required to hand over personal identifying information, e.g. passport details, or driver’s licence, to access websites or being subject to behavioural profiling.

The report complains that this imposes “cumbersome” duties on more than 25,000 companies, including ‘Big Tech’ brands such as Google and Facebook, web forums such as Mumsnet, and collaborative sites like Wikipedia. .

The duties risk being disproportionately costly for smaller, challenger firms, undermining competition and innovation. The briefing warns this could replicate the unintended consequences of GDPR, which a recent study concluded led to an 8 per cent reduction in profits for smaller firms while having no effect on the profitability of large technology companies.

The authors warn that imposing immense regulatory burdens on platforms is highly likely to lead to substantial overcompliance or, alternatively, to platforms leaving the UK.

Victoria Hewson, the IEA’s head of regulatory affairs, and co-author of ‘An Unsafe Bill’, said: “It is ironic that the government is pursuing this at the same time as it seeks to safeguard free speech in universities. Are we looking at speech being protected on campus, but censored on social media?”

Lord Frost, a senior fellow at Policy Exchange, said: “There is so much wrong with this Bill that it is hard to know where to start, but the report rightly highlights the fact that it will mean some speech that is legal offline will effectively be impossible online. That makes no sense and will be highly damaging to public debate, especially given the weakness of the free speech protections in the Bill.”

Conservative MP David Davis said that while “the Government no doubt has good intentions, in its current form, the Bill could end up being one of the most significant accidental infringements on free speech in modern times.”

Ruth Smeeth, CEO of Index on Censorship and former Labour MP, said: “The implications of the bill also push beyond our borders. At Index on Censorship, we work hard across the world fighting for freedom of expression in some of the most hostile environments. This bill will make it harder for freedom fighters to push back globally.”

A DCMS spokesperson said: “The bill will not stifle online rights and freedoms. While in-scope companies will need to tackle criminal activity and protect children, nothing in the bill requires them to remove or moderate legal content accessed by adults.

“The bill includes tough and targeted safeguards for free speech, political debate and journalism. We will ensure that users can appeal the wrongful removal of content.”