Online Safety Bill

The UK's disgraceful Online Safety Bill has passed through Parliament and will soon become law. The wide-ranging legislation, which is likely to affect every internet user in the UK and any service they access, and generate mountains of onerous red tape for any internet business stupid enough to be based in Britain. Potential impacts are still unclear and some of the new regulations are technologically impossible to comply with.

A key sticking point is what the legislation means for end-to-end encryption, a security technique used by services like WhatsApp that mathematically guarantees that no one, not even the service provider, can read messages sent between two users. The new law gives regulator Ofcom the power to intercept and check this encrypted data for illegal or harmful content.

Using this power would require service providers to create a backdoor in their software, allowing Ofcom to bypass the mathematically secure encryption. But this same backdoor would be abused by hackers, thieves, scammers and malicious states to snoop, steal and hack.

Beyond encryption, the bill also brings in mandatory age checks on pornography websites and requires that websites have policies in place to protect people from harmful or illegal content. What counts as illegal and exactly which websites will fall under the scope of the bill is unclear, however.

Neil Brown at law firm decoded.legal says Ofcom still has a huge amount of work to do. The new law could plausibly affect any company that allows comments on its website, publishes user-generated content, transmits encrypted data or hosts anything that the government deems may be harmful to children, says Brown:

What I'm fearful of is that there are going to be an awful lot of people, small organisations - not these big tech giants -- who are going to face pretty chunky legal bills trying to work out if they are in scope and, if so, what they need to do.

The U.K.'s Online Safety Bill has passed a critical final stage in the House of Lords, and envisions a potentially vast scheme to surveil internet users.

The bill would empower the U.K. government, in certain situations, to demand that online platforms use government-approved software to search through all users' photos, files, and messages, scanning for illegal content. Online services that don't comply can be subject to extreme penalties, including criminal penalties.

Such a backdoor scanning system can and will be exploited by bad actors. It will also produce false positives, leading to false accusations of child abuse that will have to be resolved. That's why the bill is incompatible with end-to-end encryption--and human rights. EFF has strongly opposed this bill from the start.

Now, with the bill on the verge of becoming U.K. law, the U.K. government has sheepishly acknowledged that it may not be able to make use of some aspects of this law. During a final debate over the bill, a representative of the government said that orders to scan user files can be issued only where technically feasible, as determined by Ofcom, the U.K.'s telecom regulatory agency. He also said any such order must be compatible with U.K. and European human rights law.

That's a notable step back, since previously the same representative, Lord Parkinson of Whitley Bay, said in a letter to the House of Lords that the technology that would magically make invasive scanning co-exist with end-to-end encryption already existed . We have seen companies develop such solutions for platforms with end-to-end encryption before, wrote Lord Parkinson in that letter.

Now, Parkinson has come quite close to admitting that such technology does not, in fact, exist. On Tuesday, he said :

There is no intention by the Government to weaken the encryption technology used by platforms, and we have built strong safeguards into the Bill to ensure that users' privacy is protected.

If appropriate technology which meets these requirements does not exist, Ofcom cannot require its use. That is why the powers include the ability for Ofcom to require companies to make best endeavors to develop or source a new solution.

The same day that these public statements were made, news outlets reported that the U.K. government privately acknowledged that there is no technology that could examine end-to-end encrypted messages while respecting user privacy.

 

People Need Privacy, Not Weak Promises

Let's be clear: weak statements by government ministers, such as the hedging from Lord Parkinson during this week's debate, are no substitute for real privacy rights.

Nothing in the law's text has changed. The bill gives the U.K. government the right to order message and photo-scanning, and that will harm the privacy and security of internet users worldwide. These powers, enshrined in Clause 122 of the bill, are now set to become law. After that, the regulator in charge of enforcing the law, Ofcom, will have to devise and publish a set of regulations regarding how the law will be enforced.

Several companies that provide end-to-end encrypted services have said they will withdraw from the U.K. if Ofcom actually takes the extreme choice of requiring examination of currently encrypted messages. Those companies include Meta-owned WhatsApp, Signal, and U.K.-based Element, among others.

While it's the last minute, Members of Parliament still could introduce an amendment with real protections for user privacy, including an explicit protection for real end-to-end encryption.

Failing that, Ofcom should publish regulations that make clear that there is no available technology that can allow for scanning of user data to co-exist with strong encryption and privacy.

Finally, lawmakers in other jurisdictions, including the United States, should take heed of the embarrassing result of passing a law that is not just deceptive, but unhinged from computational reality. The U.K. government has insisted that through software magic, a system in which they can examine or scan everything will also somehow be a privacy-protecting system. Faced with the reality of this contradiction, the government has turned to an 11th hour campaign to assure people that the powers it has demanded simply won't be used.

Natasha Kaplinsky, anti-porn campaigner and president of the British Board of Film Classification (BBFC), said in an exclusive paywalled article for the Sunday Telegraph, said that the Government needed not only to introduce tough age verification to protect children but also to take action to restrict young adults from accessing the welter of illegal violent and abusive porn available online.

Kaplinsky, who is also president of the UK's biggest children's charity Barnardo's, is calling for amendments to the Online Safety Bill, currently before the House of Lords, that would bring the legislative treatment of porn on the internet in line with the restrictions that the BBFC polices in the offline world.

Kaplinsky cited a recent parliamentary report which revealed illegal porn was readily accessible online including depictions of rape, incest and sexual violence. She said:

This was because the offline regulation of legitimate porn overseen by the BBFC was not mirrored online and the Government's Bill as written did not plug this loophole. This meant content that would be illegal to distribute offline will continue to be legally available online,

She claimed that this attempt at further internet censorship was not an attempt at censorship:

To be clear: this is not about limiting the freedom of adults to access legal pornographic material.

This is about the regulation of appalling content that eroticises rape and the violent abuse of women, or which promotes an interest in abusive relationships. There is a big difference. It is only logical that where content is unacceptable offline, we as a society should say it is unacceptable online too.

Presumably her reference to promoting an interest in abusive relationships is a reference to the plethora of 'step family' porn, but it must be noted that the BBFC has passed such material R18, eg see That's Right, She's My Step Sister...so What from bbfc.co.uk

Attempts to extend the censorship of porn online are expected next month when it comes before the Lords.

The Online Safety Bill is currently going back to Report Stage in the Commons on 16 th January, and is widely expected to be in the Lords for the end of the month, or beginning of February. We anticipate it could complete its legislative passage by June.

At the end of last year, a widely publicised change to the Online Safety Bill took out the so-called "legal but harmful" clauses for adults. The government has claimed this is protecting free speech.

However, in their place, new clauses have been shunted in that create a regime for state-mandated enforcement of tech companies' terms and conditions. It raises new concerns around embedded power for the tech companies and a worrying lack of transparency around the way that the regulator, Ofcom, will act as enforcer-in-chief.

Whatever they say goes

It is not a good look for free speech. It does not alter the underlying framework of the Bill that establishes rules by which private companies will police our content. On the other hand, it does create a shift in emphasis away from merely "taking down" troublesome content, and towards "acting against users".

For policy geeks, the change removed Clauses 12 and 13 of the Bill, concerning "content harmful to adults". The clauses regarding harmful content for children, Clauses 10 and 11, remain.

The two deleted clauses have been replaced by five new clauses addressing the terms of service of the tech companies. If their terms of service say they will "act against" content of "a particular kind", then they will follow through and do so. This will be enforced by Ofcom.

The new clauses emphatically refer to "restricting users' access" as well as taking down their content, or banning users from the service. The language of "restricting access" is troubling because the implied meaning suggests a policy of limiting free speech, not protecting it. This is an apparent shift in emphasis away from taking down troublesome content, to preventing users from seeing it in the first place. It is an environment of sanctions rather than rights and freedoms.

There is no definition of "a particular kind" and it is up to the tech companies to identify the content they would restrict. Indeed, they could restrict access to whatever they like, as long as they tell users in the terms of service.

The political pressure will be on them to restrict the content that the government dictates. It will not be done by the law, but by backroom chats, nods and winks over emails between the companies, Ofcom and government Ministries.

Joining the dots, Ofcom has a legal duty to "produce guidance" for the tech companies with regard to compliance. Ofcom takes direction from the two responsible Ministries, DCMS and the Home Office. A quick call with expression of the Minister's concerns could be used to apply pressure, with the advantage that it would skirt around publicly accountable procedures. "Yes, Minister" would morph into real life.

Restricting access to content

The new clauses do attempt to define "restricting users access to content". It occurs when a tech company "takes a measure which has the effect that a user is unable to access content without taking a prior step" or "content is temporarily hidden from a user". It's a definition that gives plenty of room for tech companies to be inventive about new types of restrictions. It does seem to bring in the concept of age-gating, which is a restriction on access, requiring people to take the step of establishing their identity or age-group, before being allowed access.

The new provisions also state that tech companies "must not act against users except in accordance with their terms and conditions", but the repetition of restrictive language suggests that the expectation is that they will restrict. There is no recognition of users' freedom of expression rights, and they may only complain about breach of contract, not breach of rights.

These restrictive clauses should also be seen in light of another little twist of language by the Bill's drafters: "relevant content". This is any content posted by users onto online platforms, but it is also any content capable of being searched by search engines, which are in scope of the Bill. The mind boggles at how much over-reach this Bill could achieve. How many innocent websites could find themselves demoted or down-ranked on the basis of the government whim of the day?

"Relevant content" is applicable when users seek to complain. But how can users complain about their website being down-ranked in a search listing when they don't have any confirmation that it has happened? The Bill makes no provision for users to be informed about "restricted access".

The change fails to take account of the potential cross-border effects, that will especially affect search functions. The Bill limits its jurisdiction to what it calls "UK-linked" content or web services. The definition is imprecise and includes content that is accessible from the UK. Online platform terms and conditions are usually written for a global user base. It's not clear if this provision could over-reach into other jurisdictions, potentially banning lawful content or users elsewhere.

Failure of policy-making

It reflects a failure of policy-making. These platforms are important vehicles for the global dissemination of information, knowledge and news. The restrictions that online platforms have in their armoury will limit the dissemination of users' content, in ways that are invisible and draconian. For example, they could use shadow bans, which operate by limiting ways that content is shown in newsfeeds and timelines. The original version of the Bill as introduced to Parliament did acknowledge this, and even allowed user to complain about them. The current version does not.

Overall, this is a failure to recognise that the vast majority of users are speaking lawfully. The pre-Christmas change to the Bill puts them at risk not only of their content being taken down but their access being restricted. Freedom of expression is a right to speak and to be informed. This change affects both.

The Online Safety Bill is back in Parliament. It had been stalled for five months whilst the government made a few changes. A Parliamentary debate on Monday (5th December) revealed the shift in policy direction for the first time. It's relatively small change, with big implications.

According to the government, the Online Safety Bill is supposed to protect children. However, from a digital rights perspective it is probably the most worrying piece of legislation ever imagined to date. The government's focus is on the content it wants to ban, with little attention paid to the impact on freedom of expression or privacy. The lack of definition or precision in the text leaves wide open loopholes for over-removals of content and the possibility of government- imposed, privatised surveillance.

The emphasis was on new amendments to be tabled early next year. Self-harm content, deep fakes and the sharing of non-consensual intimate images, will be defined as new criminal offences and illegal content.

The subtle policy shift turns on a requirement for large online platforms to tackle the so-called "legal but harmful" content. This is a legally-problematic, grey area. It is about content that is not illegal but which the government wants to ban, and understood to include eating disorders, self-harm, and false claims about medicines.

The government has announced a plan to delete this requirement, but only for adult users, not for children. An amendment will be tabled next week.

A further, legally problematic, amendment requires platforms to allow adult users to filter out these kinds of harmful content for themselves. The idea is a kind of filter button where users can select the type of harmful content that they don't want to see.

In tandem, there will be an amendment that makes online platforms enforce their terms and conditions with regard to content that is not addressed by the Bill.

We have seen drafts of some of these amendments, and await the final versions.

This filter, together with the requirement to enforce terms and conditions, and an existing requirement to remove all illegal content, is what the government is calling its "triple shield". The government claims this will protect users from the range of harms set out in the Bill. It also claims the move will protect free speech. This claim does not stack up, as the underlying censorship framework remains in place, including the possibility of general monitoring and upload filters.

Moreover, the effect of these amendments is to mitigate in favour of age-gating. The notion of "legal but harmful" content for children remains in the Bill. In Monday's debate, government Ministers emphasised the role of "age assurance" which is a requirement in the Bill although it does not say how it should be implemented.

The government's position on age-gating is broader than just excluding under-18s from 'adult' content. The Secretary of State, Michelle Donelan, said that all platforms must know the age of their users. They may be required to differentiate between age-groups, in order to prevent children from engaging with age-inappropriate harmful content to be defined by the government. The likely methods will use biometric surveillance.

MPs have also passed an amendment that confirms chat controls on private messaging services. This is the "spy clause" , renumbered S. 106 (formerly S.104). It's a stealth measure that is almost invisible in the text, with no precision as to what providers will do. The government's preferred route is understood to be client-side scanning. This completes a trio of surveillance on public posts, private chats and children.

The Times is reporting that the government's Online Censorship Bill will return to the House of Commons on December 5th with a few amendments re 'harmful but legal' content.

Rishi Sunak is to introduce a compromise over the Online Safety Bill that will involve users being able to filter out legal but harmful content without it being removed by tech platforms.

The bill  has been paused while the government takes out provisions that alarmed free speech advocates. Of particular concern were sections that would have led to tech platforms such as Facebook, Instagram, TikTok and Google removing content that was deemed to be legal, but harmful to adults.

The government will also detail a new offence about sharing deep fake porm. Those who share pornographic deepfakes,explicit images or videos that have been manipulated to look like someone without their consent, could be jailed under the proposed changes. It is not clear how the government will take on the international porn websites where faked porn of celebrities is commonplace. Perhaps the government will have to block them all.

Meanwhile the censorship bill is causing further criticisms over governments powers to degrade encryption. This is used to keep British people safe from hackers, blackmailers  and thieves, not to mention snooping by malicious governments most notably China and Russia.

The Open Rights Group explains in an article from openrightsgroup.org :

The Online Safety Bill requires ALL online speech to be monitored for harmful content, including the private conversations you have on your phone with friends and family. Companies like Whatsapp and Signal will be required by law to break end-to-end encryption, so the Government can automatically scan your messages.

They say encryption is dangerous, but the opposite is true. Encryption keeps your information and transactions safe from criminals. It ensures your private messages stay private. If the UK Government can break encryption to read your messages, that means scammers, hackers and foreign governments can too. Save encryption, Protect the security of your phone

If they get their way, your phone will be turned into a spy in your pocket. Billions of personal messages will be ready to be hacked, sold and exploited. The Government's plan to access your private messages will help criminals and make us less safe.

The Telegraph is reporting on significant changes being considered by the government to its Online Censorship Bill.

The government is considering backing off from the government defined censorship of 'legal but harmful' content on most websites available in the UK. The government has rightfully been taking stick for these free speech curtailing measures, particularly as the censorship is expected to be implemented mostly by mostly woke US internet giants who clearly don't care about free speech, and will over censor to ensure that they don't get caught up in the expense of getting it wrong by under censoring.

Culture Secretary Michelle Donelan is said to be considering the option for adults to be able to self censor 'legal but harmful' content by clicking a filter button that will order websites to block such content. Of course children will not be able to opt out of that choice. And of course this will men that age and identity verification has to be in place to esnsure that only adults can opt out.

A Culture Department spokesman said:

The Secretary of State has committed to strengthen protections for free speech and children in the Online Safety Bill and bring the bill back to the Commons as soon as possible. It remains the Government's intention to pass the bill this session.

PoliticsHome spotted the change to the House of Commons schedule last night, reporting that the Online Censorship Bill had been dropped from the Commons business next week.

A source in the Department of Digital, Culture, Media and Sport (DCMS) told TechCrunch that the latest delay to the bill's parliamentary timetable is to allow time for MPs to read new amendments -- which they also confirmed are yet to be laid.

But they suggested the delay will not affect the passage of the bill, saying it will progress within the next few weeks.

The change of PM may not mean major differences in policy approach in the arena of online regulation as Rishi Sunak has expressed similar concerns about the Online Safety Bill's impact on free speech -- also seemingly centred on clauses pertaining to restrictions on the legal but harmful speech of adults.

Plans to compel social media platforms to tackle online harms are not fit for purpose according to a new poll of IT experts.

Only 14% of tech professionals believed the Online Harms Bill was fit for purpose, according to the survey by BCS, The Chartered Institute for IT.

Some 46% said the bill was not workable, with the rest unsure.

The legislation would have a negative effect on freedom of speech, most IT specialists (58%) told BCS.

Only 19% felt the measures proposed would make the internet safer, with 51% saying the law would not make it safer to be online.

There were nearly 1,300 responses from tech professionals to the survey by BCS.

Just 9% of IT specialists polled said they were confident that legal but harmful content could be effectively and proportionately removed.

Some 74% of tech specialists said they felt the bill would do nothing to stop the spread of disinformation and fake news.

Offsite Article: French lawyers provide the best summary yet

15th June 2022. See article from taylorwessing.com

 

Offsite Article: Have we opened Pandora's box?

20th June 2022. See article from tandfonline.com

Abstract

In thinking about the developing online harms regime (in the UK and elsewhere1) it is forgivable to think only of how laws placing responsibility on social media platforms to prevent hate speech may benefit society. Yet these laws could have insidious implications for free speech. By drawing on Germany's Network Enforcement Act I investigate whether the increased prospect of liability, and the fines that may result from breaching the duty of care in the UK's Online Safety Act - once it is in force - could result in platforms censoring more speech, but not necessarily hate speech, and using the imposed responsibility as an excuse to censor speech that does not conform to their objectives. Thus, in drafting a Bill to protect the public from hate speech we may unintentionally open Pandora's Box by giving platforms a statutory justification to take more control of the message.

See full article from tandfonline.com

 

Offsite Article: The Online Safety Act - An Act of Betrayal

5th July 2022. See article from ukcolumn.org by Iain Davis

The Online Safety Bill (OSB) has been presented to the public as an attempt to protect children from online grooming and abuse and to limit the reach of terrorist propaganda.

This, however, does not seem to be its primary focus. The real objective of the proposed Online Safety Act (OSA) appears to be narrative control.

The Guildford Diocesan Synod has submitted a motion to the General Synod, the Church of England's legislative body, seeking to prevent children and young people from online exposure to pornography. The General Synod, which takes place in York next month, will consider the motion.

In the papers published last week, the Rev Charleen Hollington, a member of the Leatherhead Deanery Synod, Guildford, wrote:

Access to pornography means that a distorted and harmful view of what constitutes normal sexual relations is being absorbed by each new generation of children and young people.

This is placing pressure on young boys and girls to conform to stereotypes of domination on the one hand and submission and degradation on the other, and is creating a wider culture of abusive attitudes towards girls and women.

A law requiring age verification for access to commercial porn sites was meant to come into effect in 2018, but it never did for reasons having to do with bureaucratic delay and then a changed approach by the Government. 'Increase awareness of harms of pornography'

Hollington also criticised the Department for Digital, Culture, Media and Sport's Bill, saying:

If passed, the proposed legislation will go some way to addressing the problems. However, legislation introduced in 2018 which was designed to require age verification for access to commercial porn sites never came into effect.

Therefore, the need for the motion to be passed by General Synod now remains as strong as it has always been. The motion acknowledges the current problem, asks the Government to take action and recommends programmes to increase awareness of the harms of pornography.

So what will porn users do should their favourite porn site succumb to age verification. Will they decide to use a VPN, or else try Tor, or perhaps exchange porn with their friends, or perhaps their will be an opportunity for a black market to spring up. Another option would be to seek out lesser known foreign porn sites that van fly under the radar.

All of these options seem more likely than users dangerously handing over identity documents to any porn website that asks.

According to a new survey from YouGov, 78% of the 2,000 adults surveyed would not be willing to verify their age to access adult websites by uploading a document linked to their identity such as a driver's license, passport or other ID card.

Of the participants who believe that visiting adult websites can be part of a healthy sexual lifestyle, just 17% are willing to upload their ID.

The main reasons for their decisions were analysed. 64% just don't trust the companies to keep their data safe while 63% are scared their information could end up in the wrong hands. 49% are concerned about adult websites suffering data breaches which could expose their personal information.

Director of the privacy campaigner Open Rights Group, Jim Killock explained in a press release that those who want to access adult websites anonymously will just use a VPN if the UK's Online Safety legislation passes, saying:

The government assumes that people will actually upload their ID to access adult content. The data shows that this is a naive assumption. Instead, adults will simply use a VPN (as many already do) to avoid the step, or they'll go to smaller, unmoderated sites which exist outside the law. Smaller adult sites tend to be harder to regulate and could potentially expose users204including minors204to more extreme or illegal content.

Repressive new censorship laws return to Parliament for their second reading this week.

Online censorship legislation will be debated in the Commons Comes as new plans to support some people and fight  deemed falsities online are launched Funding boost will help people's critical thinking online through a new expert Media Literacy Taskforce alongside proposals to pay for training for teachers and library workers

Parliamentarians will debate the government's groundbreaking Online Censorship Bill which requires social media platforms, search engines and other apps and websites allowing people to post content to censor 'wrong think' content.

Ofcom, the official state censor, will have the power to fine companies failing to comply with the laws up to ten per cent of their annual global turnover, force them to improve their practices and block non-compliant sites. Crucially, the laws have strong measures to safeguard children from harmful content such as pornography and child sexual abuse.

The Online Safety Bil l was published on 12 May 2021 with the stated aim of cracking down on harmful content online. A clause has now been added to the bill to include the offence of inciting or controlling prostitution for gain as one of the priority offences that tech companies have to look out for -- firms would then be obliged to remove any content from their platforms that could be construed as committing this offence.

This would be disastrous for sex workers as it would undoubtably lead to advertising platforms clamping down on sex workers' advertisements.in order to avoid any chance of being prosecuted -- essentially criminalising the online advertising of sex work.

Controlling prostitution for gain is interpreted very widely in the criminal courts. Some women in the ECP have been prosecuted under this offence just for helping a friend build a website or place an advert. Our experience shows that in any crackdown like this, migrant and women of colour are particularly targeted.

Research shows that online advertising has enabled sex workers to work more safely and independently from exploitative bosses, to screen clients and have more control over our working conditions. Preventing sex workers from advertising will increase violence and the risk of attack. Similar legislation (SESTA/FOSTA) was passed into law by Trump in the US in 2018 resulting in an increase in poverty, insecure housing, suicide, murder, isolation, and the deterioration of physical and mental health for sex workers.

The UK government's new online censorship laws have been brought before parliament. The Government wrote in its press release:

The Online Safety Bill marks a milestone in the fight for a new digital age which is safer for users and holds tech giants to account. It will protect children from harmful content such as pornography and limit people's exposure to illegal content, while protecting freedom of speech.

It will require social media platforms, search engines and other apps and websites allowing people to post their own content to protect children, tackle illegal activity and uphold their stated terms and conditions.

The regulator Ofcom will have the power to fine companies failing to comply with the laws up to ten per cent of their annual global turnover, force them to improve their practices and block non-compliant sites.

Today the government is announcing that executives whose companies fail to cooperate with Ofcom's information requests could now face prosecution or jail time within two months of the Bill becoming law, instead of two years as it was previously drafted.

A raft of other new offences have also been added to the Bill to make in-scope companies' senior managers criminally liable for destroying evidence, failing to attend or providing false information in interviews with Ofcom, and for obstructing the regulator when it enters company offices.

In the UK, tech industries are blazing a trail in investment and innovation. The Bill is balanced and proportionate with exemptions for low-risk tech and non-tech businesses with an online presence. It aims to increase people's trust in technology, which will in turn support our ambition for the UK to be the best place for tech firms to grow.

The Bill will strengthen people's rights to express themselves freely online and ensure social media companies are not removing legal free speech. For the first time, users will have the right to appeal if they feel their post has been taken down unfairly.

It will also put requirements on social media firms to protect journalism and democratic political debate on their platforms. News content will be completely exempt from any regulation under the Bill.

And, in a further boost to freedom of expression online, another major improvement announced today will mean social media platforms will only be required to tackle 'legal but harmful' content, such as exposure to self-harm, harassment and eating disorders, set by the government and approved by Parliament.

Previously they would have had to consider whether additional content on their sites met the definition of legal but harmful material. This change removes any incentives or pressure for platforms to over-remove legal content or controversial comments and will clear up the grey area around what constitutes legal but harmful.

Ministers will also continue to consider how to ensure platforms do not remove content from recognised media outlets.

Bill introduction and changes over the last year

The Bill will be introduced in the Commons today. This is the first step in its passage through Parliament to become law and beginning a new era of accountability online. It follows a period in which the government has significantly strengthened the Bill since it was first published in draft in May 2021. Changes since the draft Bill include:

• Bringing paid-for scam adverts on social media and search engines into scope in a major move to combat online fraud .

• Making sure all websites which publish or host pornography , including commercial sites, put robust checks in place to ensure users are 18 years old or over.

• Adding new measures to clamp down on anonymous trolls to give people more control over who can contact them and what they see online.

• Making companies proactively tackle the most harmful illegal content and criminal activity quicker.

• Criminalising cyberflashing through the Bill.

Criminal liability for senior managers

The Bill gives Ofcom powers to demand information and data from tech companies, including on the role of their algorithms in selecting and displaying content, so it can assess how they are shielding users from harm.

Ofcom will be able to enter companies' premises to access data and equipment, request interviews with company employees and require companies to undergo an external assessment of how they're keeping users safe.

The Bill was originally drafted with a power for senior managers of large online platforms to be held criminally liable for failing to ensure their company complies with Ofcom's information requests in an accurate and timely manner.

In the draft Bill, this power was deferred and so could not be used by Ofcom for at least two years after it became law. The Bill introduced today reduces the period to two months to strengthen penalties for wrongdoing from the outset.

Additional information-related offences have been added to the Bill to toughen the deterrent against companies and their senior managers providing false or incomplete information. They will apply to every company in scope of the Online Safety Bill. They are:

• offences for companies in scope and/or employees who suppress, destroy or alter information requested by Ofcom;

• offences for failing to comply with, obstructing or delaying Ofcom when exercising its powers of entry, audit and inspection, or providing false information;

• offences for employees who fail to attend or provide false information at an interview.

Falling foul of these offences could lead to up to two years in imprisonment or a fine.

Ofcom must treat the information gathered from companies sensitively. For example, it will not be able to share or publish data without consent unless tightly defined exemptions apply, and it will have a responsibility to ensure its powers are used proportionately.

Changes to requirements on 'legal but harmful' content

Under the draft Bill, 'Category 1' companies - the largest online platforms with the widest reach including the most popular social media platforms - must address content harmful to adults that falls below the threshold of a criminal offence.

Category 1 companies will have a duty to carry risk assessments on the types of legal harms against adults which could arise on their services. They will have to set out clearly in terms of service how they will deal with such content and enforce these terms consistently. If companies intend to remove, limit or allow particular types of content they will have to say so.

The agreed categories of legal but harmful content will be set out in secondary legislation and subject to approval by both Houses of Parliament. Social media platforms will only be required to act on the priority legal harms set out in that secondary legislation, meaning decisions on what types of content are harmful are not delegated to private companies or at the whim of internet executives.

It will also remove the threat of social media firms being overzealous and removing legal content because it upsets or offends someone even if it is not prohibited by their terms and conditions. This will end situations such as the incident last year when TalkRadio was forced offline by YouTube for an "unspecified" violation and it was not clear on how it breached its terms and conditions.

The move will help uphold freedom of expression and ensure people remain able to have challenging and controversial discussions online.

The DCMS Secretary of State has the power to add more categories of priority legal but harmful content via secondary legislation should they emerge in the future. Companies will be required to report emerging harms to Ofcom.

Proactive technology

Platforms may need to use tools for content moderation, user profiling and behaviour identification to protect their users.

Additional provisions have been added to the Bill to allow Ofcom to set expectations for the use of these proactive technologies in codes of practice and force companies to use better and more effective tools, should this be necessary.

Companies will need to demonstrate they are using the right tools to address harms, they are transparent, and any technologies they develop meet standards of accuracy and effectiveness required by the regulator. Ofcom will not be able to recommend these tools are applied on private messaging or legal but harmful content.

Reporting child sexual abuse

A new requirement will mean companies must report child sexual exploitation and abuse content they detect on their platforms to the National Crime Agency .

The CSEA reporting requirement will replace the UK's existing voluntary reporting regime and reflects the Government's commitment to tackling this horrific crime.

Reports to the National Crime Agency will need to meet a set of clear standards to ensure law enforcement receives the high quality information it needs to safeguard children, pursue offenders and limit lifelong re-victimisation by preventing the ongoing recirculation of illegal content.

In-scope companies will need to demonstrate existing reporting obligations outside of the UK to be exempt from this requirement, which will avoid duplication of company's efforts.