Britain's Online Safety Act Comes Into "Full Force"—Its Origins, the Players Involved, the Powers Buried Within, and the Alarming Future Ahead
The truth behind one of our most authoritarian pieces of legislation yet.
The Online Safety Act has been in the works since 2017.
It all began, as so much invasive, wide-reaching legislation does, with tragedy.
In 2017, a 14-year-old girl by the name of Molly Russell started consuming dark content online. Themes of self-harm and suicide relentlessly bombarded her feed, pulling her into despair.
Soon, she fell into a pit of depression. Months later, unable to escape its grip, little Molly took her own life. Reports didn’t state specifically how she died, only that it involved “self-harm”.
Her father, Ian Russell, emerged as a fierce advocate for online safety in the aftermath, criticising how social media firms operate around engagement. His campaigning gained national attention, culminating in government response.
Theresa May’s government had already published plans to address online safety the month before Molly’s suicide. Her Secretary of State for Digital, Culture, Media and Sport, Dame Karen Bradley, had released the Internet Safety Strategy Green Paper.
Aiming to make Britain the “safest place” to be online, the paper outlined plans of a voluntary code for social media companies to tackle abuse, annual reports on harmful content and responses, and a levy on tech firms to fund awareness campaigns.
Education was also key—integrating digital literacy into schools curricula for parents, children, and caregivers alike.
The initial recommendations placed some burdens on social media firms, but they were far from draconian. Some inconvenient safeguards were outlined but ones arguably needed.
Then, the tide shifted.
By April 2019, Theresa May’s Home Office and Department for Digital, Culture, Media and Sport were involved, co-publishing the Online Harms White Paper. Now, with ministers citing Molly’s fate, the scope of their plans expanded.
It was here we first saw proposals of a legal obligation for companies to take reasonable steps to safeguard users from “illegal” content (hate speech included), underage exposure to legal content and—the big one—“harmful but legal content”.
The mandate was widened to seemingly include almost everything.
They also proposed the establishment of an independent regulator to oversee compliance, develop codes of practice, and have the authority to impose sanctions on companies failing to meet their new rules.
This is what came into force on Monday (17th March 2025), with tech companies needing to complete compulsory content risk assessments, showing how their algorithms downgrade certain content.
Failure to do so could result in fines up to £18 million or 10% of their worldwide revenue.
After subsequent draftings of the bill in 2021 and legislative amendments in parliament throughout 2022, the bill, dubbed The Online Safety Act, passed through parliament and received Royal Assent in October 2023.
Campaigners successfully pressured representatives to withdraw the “harmful but legal” provision, citing its vague and subjective nature that would have no doubt had another chilling effect on online speech.
It marked a solid win. But while attention fixated on the former, the government, civil service, and stakeholders successfully pushed through more, let’s say, insidious clauses.
New Criminal Offences
One of those was Section 179, which introduced a brand new criminal offence for “knowingly sending false information” that causes “non-trivial psychological or physical harm”.
The provision obviously intends to prevent things like cyber-bulling. What we didn’t know was that it would be used by police forces to arrest citizens for speculation. You read that right.
The story of Bernadette “Bernie” Spofforth is a case in point.
On July 29th 2024, Bernie misidentified Southport child-murderer Axel Rudakubana as Ali-Al-Shakati on X hours after the heinous attack. About a week later, Cheshire Police arrested her for “stirring up racial hatred” and “false communications”.
Now, the force genuinely did not have any evidence that Bernie “knowingly” sent false communications that “caused non-trivial psychological or physical harm”. Nor, it turns out, did they have proof of “false communications”.
Bernie prefaced her X post with—wait for it—“if this is true”. Meaning, though she categorically declared she did not know the child-murderer’s identity, she was arrested, at least partially, for speculating on it.
She was reportedly held in jail for 36 hours after “being dragged” from her home. The police eventually dropped the case.
Cutting the story short…
We went from voluntary social media codes to attempts to censor “legal but harmful” content for all (not just children), to a draconian speech law—one that allowed a seemingly politicised police force to arrest a citizen for airing rumour and arguably violate her human rights in the process.
The maximum penalty for a false communications offence under Section 179 is 51 weeks in prison, a fine, or both. Worse, as a summary-only offence, defendants lose the right to a jury trial, like Section 127 of 2003 Communications Act before it.
Instead, a lay magistrate or district judge decides their fate—without the scrutiny or safeguards a jury provides, thus, stripping away a considerable degree of due process and opening up sentences to judicial bias.
Section 181 of the Online Safety Act also creates a new criminal offence: sending “seriously threatening messages” online. While it’s framed as means to curb abuse, it goes further—criminalising threats “intended to cause fear or distress”.
This includes threats of death, serious injury, sexual assault, rape, and, notably, threats of serious financial loss or damage to property.
Death and sexual assault sure, but financial loss? What qualifies? A joke about boycotting a business? A satirical meme aimed at a politician’s donors?
If the police’s prior arrest of a veteran for “malicious communications” over a meme—an LGBTQ+ flag shaped like a swastika that allegedly “caused someone anxiety”—sets any precedent, it’s clear it’ll extend to similarly trivial acts.
The punishment for breaking Section 181 is up to 5 years’ imprisonment, an unlimited fine, or both. Serious offences are tried in the Crown Courts with juries while others are tried in the Magistrates without.
For context, in 2023, close pal of former Health Secretary Matt Hancock, hereditary peer Lord James Bethell even attempted to criminalise “vaccine misinformation” by adding it to the bill.
This is same bloke who mysteriously "replaced" his phone when questions arose about £85 million contracts he awarded to certain companies for covid tests.
Unfortunately, the overreach didn’t stop there. The political class baked in yet more punitive provisions…
Preferential Application
While independent journalists and social media commentators face content restrictions, Section 50 exempts “recognised news publishers” (i.e., mainstream media).
Online platforms are not obligated to apply their new "safety duties to content from recognised news publishers. So The Guardian, for example, won’t be subject to the same regulation.
*There's even a ‘Must Carry’ obligation that platforms must not remove or restrict their content without “due process”, where they must notify the outlet first and offer them an appeal mechanism.
Advocates of the Act contend this makes sense because mainstream outlets are already rigorously regulated by the “independent” IPSO or IMPRESS.
*The exemptions could also influence arrests and prosecutions under the new “false communication” offence.
Put simply, if a mainstream journalist knowingly posts/publishes false information that causes “harm”, he/she might not be arrested. Little citizen journalist Billy, however, doing the same on his blog or social media profile might be.
By law, mainstream media and its journalists are not exempt from criminal acts. But so much of this depends on how the police continue to interpret and enforce the law.
So far, a citizen journalist—Bernie—was arrested for speculating on the identity and background of a then-murder suspect, while mainstream journalists and campiagners have done the same, if not worse, without facing police investigation, let alone arrest.
Empowerment of Ofcom (and Starmer’s Cabinet)
Then there’s the simultaneous expansion of power for Ofcom—the government-approved regulator for broadcasting, internet, telecommunications, and postal services—paired with a reduction in its “independence”.
In February 2020, before the first version of the Online Safety Bill was published and introduced to Parliament for pre-legislative scrutiny, Boris Johnson’s government made Ofcom the enforcer of the Act.
It gave the regulator huge control over how social media operates, despite them having no prior experience regulating content on this scale.
Not only that, in its final iteration, the Act granted secretaries the authority to direct Ofcom in its regulatory functions, including influencing codes of practice.
Ofcom is called an “independent” regulator, but it is ultimately accountable to parliament and the government. It was established by the government in 2003 and reports to the Department for Science, Innovation and Technology (DSIT).
The government also appoints Ofcom’s board members, including the Chair and Chief Executive.
In short, its independence was arguably non-existent already. The Online Safety Act further crushed any glimmer that it was.
Melanie Dawes, Ofcom’s current Chief Executive, has come under heavy fire in recent years.
The former Permanent Secretary Champion for Diversity and Inclusion has been accused of bias regulation, including imposing inconsistent and selective fines and unfairly dismissing complaints.
In 2022, Dawes and co fined GB News for comments made by a guest on presenter Mark Steyn’s show about the Covid response and pre-Nazi Germany.
When other hosts and guests, like LBC’s James O’Brien, directly compared Donald Trump to past fascist leaders, on the other hand, Ofcom was no where to be seen.
Another famously unbalanced ruling: Toby Young and Laura Dodsworth lodged complaints later that year over Sky’s partnership with the Behavioural Insights Team (BIT)—part owned by the UK Cabinet Office.
Sky and BIT’s collaboration in 2021 aimed to subtly “nudge” viewers, targeting kids, to back the government’s Net Zero agenda, a tactic Young and Dodsworth argued broke Ofcom’s rules.
Four months later, Ofcom dismissed the complaint, deeming climate science “broadly settled,” suggesting psychological manipulation is acceptable, as long as the message aligns with the “scientific consensus”.
It might come as little surprise that Dawes has also worked with The Patchwork Foundation, a group "focused on communities and individuals that are traditionally underrepresented".
Dawes can not only set and alter legally binding codes of practice (within a certain scope) that social media platforms, websites, and even search engines must follow, but she can fine them heavily for breaking those codes.
Put plainly, censorship powers have been concentrated in the hands of a single, seemingly compromised career civil servant. She interprets the rules, amends, and enforces them.
And if the government is unhappy with Dawes’ enforcement style, Starmer’s partisan secretaries can step in by law—potentially pressuring her into imposing regulations that align more with their agenda.
Empowerment of Police Chiefs
Of course, this doesn’t include the powers granted to police chiefs, who can selectively arrest citizens on suspicion of “knowingly sending false communications” causing “non-trivial psychological or physical harm.”
Or, indeed, arrest those they suspect of sending “seriously threatening messages” online “intending to cause fear or distress”.
Given Bernie’s case and the recent testimony of arrests and police intimidation following the Southport protests and riots, it constitutes yet another subjective and restrictive speech law that politically-captured police forces can use to punish almost anyone, for anything.
Don’t get me wrong. There are some positives.
The Act does target online content that commentators across the political spectrum agree should be removed, including child sexual abuse material, terrorism-related content (thought this could be contentious), revenge porn, non-consensual intimate images, and the promotion of self-harm or suicide.
But the main problem: it extends to “hate crimes”—a concept that our government and policing establishment have repeatedly shown they’re incapable of enforcing impartially and without fundamentally undermining free speech rights.
The vague terms, such as “harmful”, “democratic importance”, and “reasonable steps” peppered throughout the Act compounds the problem.
Put another way: the authoritarian powers it equips the government with arguably outweighs any benefits proponents claim the Act offers.
Subversion of Privacy Rights
Privacy rights have also been dented.
Section 122 forces online platforms to deploy technology that detects and removes illegal content, even within end-to-end encrypted messages.
End-to-end encryption ensures only the sender and recipient can read communications. Content scanning within these channels will inevitably weaken encryption.
As a result, users face greater risks of cyber threats, hacking, and unauthorised surveillance—the same dangers Labour Home Secretary Yvette Cooper’s recent successful push to access Apple users’ iCloud data enabled.
Elsewhere, platforms must implement technology to verify every user’s age, thereby effectively “age-gating” websites and setting the stage for more invasive verification checks.
Disproportionate Effects
Finally, we come to the issue of proportion.
The Act imposes such sweeping obligations on online platforms, covering any “user-to-user service” that allows user-generated content to be seen by others, that smaller platforms, websites, and blogs have already been shutting down.
From cycling to gaming, birdwatching to history, various UK-based online forums are disappearing. They simply can't afford moderation teams, legal compliance, or even the time to navigate the Act’s vague language.
The law is so broad that many forums—both in the UK and beyond—are switching to read-only mode or geo-blocking UK users entirely. Compliance isn’t a matter of difficulty. It’s impossible for them.
In January, the Labour government passed an amendment setting user number thresholds to determine which websites, search engines, and platforms fall under the Act—but the details remain unclear. We have to wait until summer for Ofcom to spell it out.
Wider Context
Last November, in the wake of the frenzy of questionable and prompt guilty pleas and mass sentences for non-violent Southport protestors, Labour Science Secretary Peter Kyle announced tougher enforcement of the Online Safety Act.
He and his department issued a statement on ‘Strategic Priorities for Online Safety’ where they referenced the word "robust" 3 times in one paragraph about the “vast amount of misinformation and disinformation”.
Weeks before, Starmer declared: "we’re due sentencing for online behaviour... whether you’re directly involved or whether you’re remotely involved, you’re culpable, and you will be put before the courts if you’ve broken the law."
My two cents…
The Online Safety Act gave Labour the perfect legal framework to supercharge their invasive, controlling, big statist—that much is obvious.
The Act’s origin story is less one of a reasonable response to child safety concerns than it is of how the British establishment will exploit tragedy to no ends in their painfully apparent bid ascertain complete control over our lives.
They did with Iraq. They did it with the Public Order Act 2023. They did with the National Security Act 2023. They did it with the Investigatory Powers Act 2016. They did it with the Online Safety Act 2023.
How do we go from heartbreaking teenage suicide to arresting a woman for speculating on a child-murderer’s identity for Pete’s sake? Some would say because it was never really about child protection after all.
And they’d probably be right.
Do you think if things stay the way they are our best days are ahead of us?
Are you going to sit down and watch as our media/government officials push for yet more invasive, authoritarian, censorious policy?
If you want to do something today to help change that, you can opt for a paid subscription and help citizen journalists like me attempting to stop it…
Coming soon to every country near you.
This is how you shut down the "unshutable" internet.
They knew when they released it that the internet could actually be something that they could not control, despite lots of preexisting gates already in place, giving a very sanitised version. Hemce the boogedy-boo narratives of "dark web" spread to the masses. Not saying there's not bad $#!@ available there, but like real crime, it's massively distorted in the media! But they never dreamed how disruptive tech wise even the sanitised version could get, because it connects people. People are way harder to manipulate when connected in groups, in face to face, but even online it's still hard. But they figured out by 2012 that they could shut it down through censorship, and "social safety" laws.😐
The future internet will be basically marketing/news 24/7. All they have achieved is pushing more people into underground or back into person to person meetups.🤦♀️😐
Really difficult one this.
Most right minded people have been screaming at the top of their voices for someone to do something about the vile and twisted material that these companies have been pushing/allowing to appear online.
However, as we see with so much that our politicians and civil servants do, what starts as a piece of common sense law on day one, ends up as a of piece of gold plated legislation and a nightmare for those subjected to it.
We saw this happen time and time again with various pieces of legislation from Brussels.
Red top tabloid loved to ridicule the EU for imposing yet more ludicrous and pointless rules on the British public.
In truth it was our own government taking routine guidance, or clarification of existing legislation applying to all 28 EU countries and gold plating it, turning it into a major burden for British businesses in the process.
My experience of this relates to waste and a fire retardant chemical used in foam.
At one point councils were close to either ceasing their domestic collection service for white goods such as fridges and freezers. They would now need a separate vehicle from all other domestic goods, due to the insulation foam they contained not being sealed in. Apparently it gives off some sort of gas that can contaminate materials such as soft furnishings.
Some bright young thing at DEFRA had read a report and written their own report to some other less bright and not so young thing, called a government minister.
This culminated in DEFRA sending out a consultation document which set alarm bells ringing within every council and waste management company in the UK. luckily, the results of that consultation saw the whole issue moved to the pending tray.
So now we have a draconian online safety act. This act should be protecting the most vulnerable people in our society; our children and stopping some of the more gullible and impressionable from getting sucked into seeing and doing bad things.
Instead, a shadowy left wing elite group of control freaks, seeking to suppress anything that threatens to disrupt the flow of cash into their coffers, have decided to highjack this act and gold plated it for their own purposes - suppression of criticism and pushback.
Reform better add this one to their list of things to be shredded in the first 100 days.