I am from the UK, London. I host a celebrity forum.
Here's my two pennies ole chap.
I'm not a lawyer, this isn't legal advice, etc but I am a certified G.
The narrative here is that the new law is intended to keep people safe. We should applaud the UK government for safeguarding us online. It's a great law. The UK is finally prioritizing our online well-being—though the same level of concern is yet to be seen on our streets.
Unfortunately, you're correct. In it's adolescent form, this law does not safe-guard smaller websites owners from prosecution. This has inadvertently resulted in smaller forums from closing (or thinking of closing #lfgss). I would even go as far as saying this law has killed off newer British communities from forming online.
Some may even argue, that this law (in its current form) has ultimately supressed our 'freedom of expression' and impeded our 'freedom of speech'.
I'd like to think that a responsible website owner looks after their community and stops harmful, unlawful content from being published. I will continue to do this. I don't need a law dictating 40 safety measures telling me this.
With that said, I do enjoy the thought of humouring this law.
Source Theregister - Tue 14 Jan 2025 // 12:15 UTC
Who has to publish a summary risk assessment on their website? Sites with 34 million or more monthly active UK users. The threshold is lower for services that allow users to forward or reshare user-generated content and use a content recommender system: 7 million or more UK users. Organizations with a smaller number of users will simply have to have the document ready should Ofcom request it.
As GreXXL mentioned it's mostly for large websites. However if you're a small fish, it's important that you're prepared as can be.
How to comply (the essentials of what you need to do):
My forum falls under <7m users. So I've just completed the Ofcoms checker tool (as some websites are exempt from this law) and completed Ofcom's Website risk assessment form.
Firstly, complete the Online Safety Act checker tool to see if this law applies to your website.
If it results in "Yes, you must comply" - Complete Illegal content duties record-keeping template (DOCX, 3.02 MB) located on this page (just in case the hotlink to the docx file dies).
Keep a record of this and review it annually.
The new law is being phased in tranches, so it's worth subscribing to Ofcoms Email update. Just choose that you only want updates regarding "online safety law". 😉
That's 'really' all there is to it ...for now.
Worse case scenario: Ofcom may ask you for a copy of your website's "Risk Assessment".
What made the form filling even quicker is I used Deepseek's and OpenAI's deep-thinking ability and I was done in like a few hrs. To save you time I've added an AI prompt template below:
I need help. UK has passed a law. The UK Online Safety Act 2023. So i need to complete risk assessment which outlines 17 critera's for asssessment. [✏️ EDIT THIS PART: My forum is called, it's located at "URL address", and add very detailed description of your forum]. It uses a flarum forum CMS system.
The criteria is a risk assessment against "✏️EDIT THIS PART: E.G "Hate" (CHANGE ACCORDING TO VARIOUS CRITERIA THE FORM IS ASKING YOU TO ASSESS, THERE'S LIKE 17 ON THE FORM)". The risk management enforceable by law by uk governing body Ofcom. help me write a risk assessment for first criteria.
Example:
Risk level:
[Evaluate the likelihood and impact of each kind of priority illegal content to assign a risk level (high, medium, low, negligible). Consider the relevant evidence to inform this judgement. You may consult the risk level tables found in Ofcom’s Risk Assessment Guidance (this is also covered in Step 2 of our Check how to comply with the illegal content rules)]
Risk factors considered:
[List any relevant and specific risk factors which relate to this kind of illegal harm (such as “messaging services”, “child users (under-18s)”, “user groups” etc.) from Ofcom’s Risk Profiles. Our ‘Check how to comply with the illegal content rules’ tool presents the associated kinds of illegal harm for each risk factor you select, or you can consult table 9.1 in the Risk Assessment Guidance.]
Additional characteristics considered:
[List any additional characteristics of your service which may be relevant (including user base, business models, functionalities, governance, and systems and processes) you have considered alongside the risk factors identified in Ofcom’s Risk Profiles.]
Existing controls considered:
[If you have considered the role of any existing controls already in operation on your service at the time of the risk assessment, you should record what these controls are, what risks they are intended to mitigate and how they do this, and how the consideration of the existing controls has impacted the risk level you have assigned to a kind of illegal content.]
Evidence:
[A list of the evidence, and summary of the reasoning, that has informed the assessment of likelihood and impact of this kind of priority illegal content, including any core and enhanced types of evidence.]
Footnotes
Deepseek was the best but the results often took me 40 secs to generate, most of the times it returned "sever is busy".
I could only use deepthink twice on two different accounts. It cud be for censorship reasons as topics are riske. but open AI was just dishing it out like the crap it is. Be sure to use OpenAI's reasoning model.
Regarding the Risk Assessment form, it was hilarious to read. Honestly, I've never had so much fun filling this form out. I couldn't stop laughing. It was asking how I'd stop things like 😂 (Used an image so discuss doesn't run into censorship issues):
To which my responses could have been "If I find any offences of this serious nature, I will solve it by reporting it to the police. 😂 That's why I pay my taxes. I am just an average joe, doing my 9-to-5 and on the weekends, I crack open a can of beer and browse the internet in my stained underpants. I am not MI5, Border force Patrol, Vet, Therapist or a Medical professional. I take this form as seriously as UK does with its immigration policy... 100% seriously."
Ofc I could have wrote that but I kept a professional tone using AI. 😉
