Options for the UK is a home for new ideas and radical thinking on policy from Nesta, the research and innovation foundation.

We are becoming more fragmented and polarised. You can see it in our politics, you can feel it in where and how we consume media. At the same time, the number of problems that we face as a country can feel as if it is growing, while the list of obvious solutions shrinks - particularly as budgets tighten and growth is scarce.

The public is increasingly impatient for change, in both their living standards and in public services. Governing in this landscape is hard. It's hard to find consensus, and harder still to build it where it doesn't exist. But we believe opportunities for solving big societal challenges do exist. And we think that the energy for change can be channelled towards a more positive version of the future.

Options for the UK provides ideas for how. We put forward new - and at times, radical - policy options. We work with governments to improve the ‘how' of policy, designing methods that support better and more effective delivery. We build products that enable people to engage with some of the difficult trade offs and choices that governments have to make. We also convene networks of those working in policy to together test new ideas, find out what works and strengthen our thinking. Non-partisan, we work with all three governments of the UK.

www.nesta.org.uk

Introduction: A national opportunity for digital sovereignty

Our digital public sphere urgently needs our collective, democratic attention. Platform power, trust in media and fears over online harms point to a need and an opportunity for the country to reclaim democratic control over our information environment.

We have the institutions that can bring this about – they need to be given the freedom and encouragement to do so. Foremost the BBC, whose Reithian heritage and continuing ability to convene national conversations can be at the centre of how Britain shapes its digital future. Its historic role as orchestrator of public discourse, which came naturally in the broadcast era, has been fractured by the rise of platforms in cyberspace that have separated out media content from the stage on which national conversation takes place. In cyberspace, our information environment is at least as much determined by, for example, Youtube's recommendation algorithms, moderation policies and app layout as any content-maker who posts to Youtube. There is no neutral "channel", and the periphery of form has become as important as the heart of content in the creation of the epistemic air we breathe. The moment has come to reunite these elements through purposeful innovation.

This paper proposes that the BBC should launch and lead an ambitious programme of digital experimentation to prototype and demonstrate what a properly democratic cyberspace could look like. It should involve the best of the British institutions that could support the endeavor, drawing on public bodies like the Turing Institute and foundations like Nesta. Successful innovations should be given the room to scale, even where this steps on the toes of market solutions. This represents a strategic opportunity—both for the BBC to evolve its public service mission for the digital age, and for Britain to pioneer sovereign alternatives to the platform-dominated status quo. By moving beyond content creation alone to help shape the very infrastructure and rules of digital engagement, the BBC can be true to its Reithian purpose while being at the centre of bringing to the country a solution to one of our most pressing problems.

The stakes extend far beyond any single institution. Britain requires a constructive alternative to the extractive, polarising dynamics of marketised platforms—one that demonstrates how digital spaces can serve democratic purposes rather than undermining them. The UK possesses unique assets for this challenge: world-class research institutions, a robust public service tradition, innovative digital sectors, and in the BBC, a trusted convener with unmatched reach. These capabilities, properly orchestrated, could contribute not just to Britain's digital sovereignty but to global understanding of how to civilise cyberspace.

This paper argues that the transformative public interventions that shaped Britain's media landscape in the radio and television eras offer crucial lessons for the digital age. The creation of purpose-driven institutions that operated on different principles from purely commercial actors demonstrably improved the entire media ecosystem, creating public goods while reducing public harms. The question is not whether such intervention is needed in digital spaces – the evidence of market failure is overwhelming – but how to design interventions that are both effective and appropriate to the complexity of modern media.

Media ecosystems shape where our collective attention lands and how societies understand themselves. The formation of culture, knowledge, and social cohesion occurs through complex processes that cannot be reduced to simple bilateral market transactions – the sense we make of ourselves, how our creativity and character are expressed, the directions that our culture takes, arise from a complex web of influence and interconnection that is replete with what economists would think of as "external" to the bilateral relationships of laissez-faire markets. While this paper employs the discipline of economic analysis and acknowledges market mechanisms, it recognizes that the "epistemic environment" generates such profound externalities that market solutions alone are insufficient without substantial public innovation and assistance in scaling.

The nature of that assistance is important to underline. The paper argues that the regulatory fashion of the last 15 years – to look for incentive regulation of market operators to do the job – will not be effective for such a complex public good. This was a system based on "the market knows best, but sometimes needs to be corrected at the edges”. That regulatory style might work for relatively simple goods like the provision of clean water or clean power – although even there the model is under severe strain. However, a system quite as complex and fundamental as the epistemic and cultural environment is one in which direct construction, rather than indirect correction is needed. The BBC over its entire history can be viewed, from a regulatory perspective, as a constructivist intervention in the market. Its huge success over the past 100 years points to how the country should view the intervention in cyberspace over the next 100.

The paper explores how the BBC, working with partners, could take a more active role in demonstrating alternatives for digital public spaces – prototyping new approaches to platforms, governance models, and user participation that serve public rather than purely private interests. It also proposes paths to scaling and considers funding models for this. It acknowledges the legitimate concerns about risk and resource allocation. This approach protects institutional reputation while creating space for the bold thinking Britain's digital future demands.

Historical parenthesis – Radio in the US and UK in the 1920s and 1930s

In media, what you see is what gets funded.

A very good example of this general maxim comes from a comparison of the development of the US and UK markets during the early years of radio (where it is about what was heard, rather than seen, of course).

The US had an advertising-funded model from the start. NBC and CBS established themselves as nationwide networks of broadcasters with syndicated content funded by big-brand advertising. There was also a large number of small local broadcasters that would sell airtime to the highest bidder – sometimes the national broadcasters, but often local advertising-funded content. Programmes competed for attention, developing the content that would generate the largest market shares. (They developed survey-based methods for quantifying the attention they gathered in order to sell the advertising around programmes – this was the origins of AC Nielsen.) To a first approximation, ad-funded media is a little like direct democracy: the programmes that get enough “votes” (ie, audience) are the ones that get made.

In the 20s and 30s, national networks in the US sold advertising to corporations with national brands – Procter and Gamble famously gave a name to the “soaps" – the episodic stories loved by potential customers in homes across the country. But not all programming was as innocent as soaps. By 1931, one of the most profitable programmes on CBS were the virulent, pro-Nazi antismetic broadcasts by Father Coughlin. Gerald Smith, a white-supremacist baptist with similar views, also became a radio sensation. Charles Lindbergh, already a celebrity, gathered huge radio audiences with his pro-Hitler message. Radio soon discovered that angry, polarised politics and conspiracy theories made for large audiences.

As it happens, Coughlin and Smith were eventually kicked off the big national networks. The American Jewish Committee complained to CBS about Coughlin's praise for Hitler and Mussolini, his attribution of responsibility for the depression to "international Jewish bankers”. There was the start of an advertising boycott. However, Coughlin and Smith were already celebrities and audiences were hungry for more of that fare. The shock-jock preachers pivoted to buying airtime directly from smaller local stations and developed direct-fundraising business models, asking for donations to be mailed to their churches. They continued to gather huge audiences – Coughlin was apparently reaching one quarter of US homes in the mid-30s – and huge donations.

Radio developed very differently in the UK. John Reith was appointed to be the head of the British Broadcasting Company in 1922, and by 1926, the Crawford Commission advocated the creation of a monopoly established by Royal Charter. The BBC was born. The establishment of a radio monopoly was not a technological requirement. Rather, it was to ensure that the technology “never be used to cater for groups of listeners, however large, who press for trite and commonplace performances”. Reith himself, beyond his famous establishment of the purpose of the broadcaster as being to “inform, educate and entertain", also was clear that 'It should bring into the greatest number of homes all that is best in every department of human knowledge, endeavour and achievement.'

By 1931, the BBC had its own preacher sensation – W.H.Elliott (father of the groundbreaking 1960s theatre director, Michael Elliott) who was given a Thursday evening slot to broadcast from his church of St Michael's, Belgravia. His sermons talked about the hardship among the poor, caused by the economic crisis. When, in 1931, it was suggested the broadcasts might stop for August, Elliott finished his last sermon with a message: “If you would like me to continue, why not write a postcard to the BBC"? The BBC was inundated with messages, just as Father Coughlin was when he asked for funds for his ministry. Elliott would go on to be a lasting celebrity, filling the Albert Hall with his socially conscious yet moderate sermons.

The Reithian BBC, of course, did not make Britain immune from fascism in the public sphere. Coughlin-like content was present in UK tabloids – famously, for example, with Lord Rothermere's “Hurrah for the Blackshirts!” Daily Mail leader in 1934. However, it was kept off that most entrancing of mediums – broadcast.

The Daily Mail had a circulation of around 2 million, and many of those will have read it for sports and entertainment rather than for politics. Moreover, it is not as if Coughlin, Smith, Lindbergh and their less well-known peers brought fascism to the USA. And the new public sphere created by radio in the USA was used to great effect by those who opposed facism – most famously perhaps in Roosevelt's own "fireside chats" which achieved double the audience of Coughlin's fire-and-brimstone. Nevertheless, an observer of the political economy of polarisation at that time would find a certain familiarity in today's online media. And Coughlin, Lindbergh et al should be seen as representing a risk to a country's response to the challenges of the 1930s that should be taken very seriously.

Competing philosophies of free speech – the USA vs the UK

The contrasting experiences of radio in the 1930s points to a fundamental socio-political difference between the UK and the USA in respect of their attitudes to freedom of speech. Broadcasting had natural monopoly characteristics in both places because of the need to allocate scarce spectrum to non-overlapping broadcasters, and the state therefore almost inevitably has some say in how the communication medium evolves. Actual laissez-faire in broadcasting is a Hobbesian state of nature, with whoever has the most powerful antenna able to drown out others and force their signal onto receivers. The First Amendment, borne of a society attached to its roots in the escape from state-imposed persecution of religious belief, hardly tolerates state intervention in the content of speech. The mirror image of that US history - Britain as the originator of the persecutions the early American colonists were fleeing, the one with a tight fusion of Church and State – is the political culture out of which sprang the BBC. (Frank Field once described the BBC as the inheritor, in the 1920s and via the English Idealists of the 1880s, of the Church of England.)

In the USA, the expectation of the public sphere is that it should be like the Frontier – a place of clear and stark choices, where groups can self-organise, and in which the individual can always resort to “exit” when their “voice” goes unheard. Frank Fields' point about the BBC and the Church of England (Reith was the son of a cleric and famously pious) is that in the 1920s, there was nothing controversial in the thought that the state should be involved in organising a public sphere in order to bring into every home all that is “best in every department of human knowledge, endeavour and achievement."

While spectrum scarcity required regulation, that regulation was compatible with many different views of the state's role in shaping the public sphere and what the spectrum transported. As Asa Briggs and Peter Burke write in their magisterial “Social History of Media" of the foundational thinking around the BBC:

[T]he official Crawford Committee, appointed in 1926 to inquire into the future of British broadcasting, agreed with Reith's line of thinking that monopoly was a matter of mission more than of technology. It was not the response to spectrum scarcity. While conceding that 'special wavelengths or alternative services' might provide an escape from what it called 'the programme dilemma', the committee trusted that they would 'never be used to cater for groups of listeners, however large, who press for trite and commonplace performances'. In 1927, when the recommendation of the committee to set up a public corporation by Royal Charter was implemented, the new corporation was hailed by the Fabian socialist W. A. Robson (1895–1980) as 'an invention in the sphere of social science no less remarkable than the invention of radio transmission in the sphere of natural science'.

The tension between American and British political traditions in their conception of the public sphere has recently come to the fore with JD Vance's and Elon Musk's desire to see a US-style conception of free-speech make itself undetachable from use of the Californian platforms that have conquered most of cyberspace. American technology products, American defence umbrellas and the American conception of free-speech should come as a package, they argue. The virulence of Trump's attack on the BBC – whatever one thinks of the fateful edit – should obviously be seen in this light.

In the British political tradition, there is a recognition that speech is made freer where the arena of speech is ordered by rules. The conventional rules of debate in Parliament are a very powerful model for the conduct of speech in Britain. “Order”, intoned in that particular chanting way, is recognised immediately across classes in the country. Jurgen Habermas and Hannah Arendt, two towering figures of 20th century political philosophy, have emphasised the importance of rules of conversation – the constitution of the public sphere – in the creation of conditions of political equality, itself, in their view, a condition of political freedom.

In Britain, we are very comfortable with the idea that freedom and equality are enhanced through a collective exercise in the creation of some degree of order. Of course, none of this means that the questions of what order, or how achieved, are politically settled. They are certainly not, as the politics of media policy in the UK attest. However, what is undeniable is that the “public interest" is a concept at the heart of UK media policy where “free speech” takes up almost all that space in the USA. And in cyberspace, we have de facto adopted a US position despite our political tradition that would make us more comfortable with a public sphere shaped by rules of fairness, not just freedom. This paper argues that we need to return to that tradition in cyberspace and provides a practical set of public policy interventions to do so.

Online media regulation as correctives to laissez-faire after 2010

We are familiar today with the purported downsides of online media: polarised politics, teenage mental health, reduced attention-spans, screen addiction, perhaps even reduced IQs, as well as direct harm through CSAM and content that incites violence. Many of these harms are plausibly linked directly to the online advertising business model combined with low barriers to entry as a content creator, as well as the very light content regulation implied by the platform safe harbour that means platforms are not treated as publishers when they host user-generated content. Advertising revenues follow time spent on screen, and media platforms have been adept at creating profit-maximising algorithmic feeds which keep the right users in the right state of mind to be susceptible to the right commercial message.

Polarising political messages are associated with attention-retention. It has turned out that many feeds and sites promoting extremist politics have been run as for-profit ventures by people outside the country being targeted who have no interest in the political messages themselves. Meta's profits derived from teenage vulnerabilities are now well-known: for example, they offered to target beauty-product advertisements to teenagers whom they had spotted were deleting selfies. And Meta makes multiples more out of hosting advertising from fraudsters than it ever has to worry about paying in fines. All of these harms follow the same logic of radio in 30s America: it is privately profitable to create the harmful content, and laissez-faire will therefore lead to its proliferation.

We have made some progress in updating our laws to try to halt the most egregious cases of harm from online media – the UK's Online Safety Act gives Ofcom extensive powers to investigate and fine media companies for the clearest cases of harms to children.

For all the good intentions of that legislation, it comes out of a style of regulation that provides only a very partial answer to the problems at hand. The Tory governments of the 2010s had a sense that there were important market failures in online media beyond the egregious harms, but their approach tended to be one of "making markets work" rather than an enthusiastic embrace of non-market alternatives. Indeed, the general narrative of the times was that bandwidth scarcity was the fundamental reason that broadcast media had tended to natural monopoly, and that this had to be regulated as a normal natural monopoly problem. ITV, for example, was given public service broadcasting obligations in exchange for the "license to print money" represented by ad-funded terrestrial broadcasting. As bandwidth scarcity disappeared – in the UK, at first thanks to satellite, then thanks to broadband and mobile – the dominant narrative was that media was becoming a "normal" market and consumer choice could do all the regulation that was needed. This narrative, of course, advantaged those, like Murdoch, sitting on the largest subscriber base for satellite television, and he made no secret of his dislike of competition from the BBC.

The idea that markets and consumer choice could solve the problems of media were reinforced by the proliferation of media in cyberspace, despite the clear forces of concentration at work in the platform economy. For example, the Furman Review (2018), which led to the establishment of the Digital Markets Unit at the CMA had as a guiding philosophy that the market power of platforms can be fixed by more competition: even if too much profit in the advertising value chain stayed in the hands of Google, Meta and Apple rather than going to consumers or content-makers, there was nothing that competition could not ultimately fix. The Cairncross Review into the platforms' impact on journalism (2018) went a little further in its critique, arguing that certain types of journalism – especially investigations and reporting relevant to local politics – are a public good, creating social benefits beyond their private ones. For example, the review acknowledged the “scarecrow effect" by which the presence of journalists helps to keep politics clean, and the serendipity effect by which someone reading a local paper for classified ads might become informed about local political matters.

However, in the same breath, the review recommended that the BBC's coverage of local news be investigated to ensure that it did not “crowd out” commercial alternatives. Music to the ears of the old press holdings, happy to dish out low quality ambulance-chasing local “news” in the space left by the BBC. What is more, while the Cairncross Review was more ambitious and interventionist than the Furman Review, it was only the latter that was actually translated into policy by the Tories.

What was never seriously considered by the last administration was any intervention similar to what we had with radio in the 20s and with TV in the postwar: a concerted public sector effort to create a media through democratic choice rather than simply accept and tweak the emergent outcome of a market-based process. Indeed, throughout the 2010s, Ofcom was pressured to limit BBC innovation, especially online, under the argument that it should not threaten the commercial revenue streams of competitors. After the 2016 Charter Renewal, Ofcom was explicitly tasked with conducting Market Impact Assessments of proposed new BBC services. The Murdoch media properties regularly argued that BBC innovation was crowding out their own commercial activities and should be stopped. They never failed to argue against the license fee and in favour of a subscription fee, like Sky's. The News Media Association objected to unpaywalled BBC online journalistic content on the same grounds.

In the early radio and TV eras, the BBC had created a positive alternative to what the market alone would have made. This positive intervention certainly limited the scope for commercial rivals – radio in the 20s and 30s is a very clear example. Commercial providers did find niches – independent, ad-funded radio proliferated from the 1960s, and ITV, Channel 4 and Channel 5 were able to sustain free-to-air ad-funded TV broadcast businesses despite the BBC's presence. The cable and satellite operators could also grow successful businesses.

However, as the Murdoch empire never ceased to argue, every attention-minute that the BBC managed to capture was one that could otherwise have gone to a commercial provider who could have extracted some quantum of profit from selling advertising or subscriptions against that attention. The presence of the BBC changed the dynamic of market competition – and made it much harder for the commercial operators, both by setting a high bar for programme quality and by restricting the attention minutes available to ad-funded products.

Public policy towards media was thus fundamentally different in the radio and TV age to what it has been in the online age. In the broadcast technologies, a positive public intervention did a huge amount of market-shaping work. In the early days of the web, BBC Online, the website shaped Britain's experience of the web, especially in Sports and Local News. However, when the web really turned towards a media driven by user generated content – first in the form of email & Usenet groups, then social media – the BBC lost its confidence. This was through a combination of internal choices and external pressures.

The BBC's retreat from the frontier of media consumption was reflected in central government's attitude over the last 25 years. While broadcast media continued to be positively regulated under public interest and public service criteria, cyberspace was left more or less to its own devices. The spirit of the times seems to have taken quite literally the call in the Grateful Dead's John Perry Barlow's “Declaration of independence of cyberspace" at Davos in 1996: “Out of our way, you old giants of atoms". That distinction continues to this day, where pornography is still dutifully censored and rated for cinema release (see the current discussion of the Crime and Policing Bill in the Lords) while everything, including the harmful and illegal, is in fact, for the savvy, a few Tor installs, VPNs and clicks away.

The Internet, it seems, really has, to date, managed to carry the conception of the public sphere as the American frontier along with the technology firms that have dominated it to date.

Going beyond the corrective – we need positive public service interventions in online media

The 2010s was the high point of the "British model” of economic regulation. In 2011, the Tory and Liberal Democrat coalition published a short triumphalist document – "The Principles of Economic Regulation” – which encapsulated their consensus view: "In the UK, economic regulation has aimed to promote effective competition where this is possible, and to provide a proxy for competition, with protection of consumers' interests at its heart, where it is not meaningful to introduce competition." This is the philosophy that leads to correctivism: the good state of affairs is that which would be produced by a competitive market, and where that cannot be achieved by laissez-faire, policy should somehow proxy the process of competition.

The Cairncross and Furman reviews acknowledged that intervention was needed and their preferred solutions were the “proxies for competition" advocated in the Principles. However, these "proxies” assume an unlikely ability for regulators to minutely determine outcomes through incentive systems. Take the example of designing the algorithm that will determine what a teen sees on their social media screen. The basic incentive of the profit-maximising firm is to maximise screen-time and deploy addictive design techniques to do so. Now imagine a purely correctivist regulator taking the problem seriously. They might, for example, fine firms for overly-addictive algorithms. But how would excess addiction be determined? A regulatory process would be established with measurement, appeals, etc. A social media firm wanting to push boundaries would find ample room to continue addicting within the bounds permitted.

Compare that with a positive, constructivist approach – one in which, instead of using the indirect lever of incentives to change outcomes, seeks to implement the outcomes directly. A corporation with a fiduciary duty to a purpose – for example, "to inform, educate and entertain” – would be tasked with developing an algorithm that fulfilled the purpose. It would develop internal mechanisms to judge its successes and would, in the best of cases, be subject to outside scrutiny and accountability in its achievement and interpretation of its purpose. Whatever the algorithm that came out of such a corporation, it would probably not be designed for maximum addictiveness within the bounds of the rules.

The point here is not to describe in detail what would be different between the two products, but rather to make it clear why one process can be expected to lead to a different sort of outcome. The correctivist regulatory process is one which fails to create a specific positive alternative because it cannot ever specify in enough detail what it is that it wants to incentivise. The constructivist version specifies in every detail the alternative, but only because it has to create it.

Domains where the definition and specification of good outcomes is very hard are ones where the correctivist philosophy will fail. This is the well-known economic phenomenon of contractual incompleteness, as developed by Oliver Hart, who develops the theoretical models showing that in order to achieve acceptable outcomes, where contracts in some important sense cannot minutely specify the requirement, it is necessary to give residual decision-making power to bodies whose incentives are most closely aligned to those of the public.

The "British Model” has recently been shown not to have performed very well even for such apparently simple cases as the provision of clean water or the building of energy infrastructure. These are goods that are in principle much easier to measure and target than “addictiveness” or “truthful news” or “constructive disagreement”, and even there the regulatory model is seriously stretched. Therefore, it should be no surprise that the incentive-regulation model of the 2010s is inadequate to the complex task of building an epistemic environment. The central claim of this paper is that the BBC, with help from other institutions, should take on this task, and today, that means rebuilding the digital public sphere.

The prehistory of social media at the BBC – the Community Programme Unit

In broadcast media as envisaged by the BBC, the broadcaster largely controls both form and content and is responsible for both. Every aspect of what is presented to attention must be answered for, and that is where the broadcaster's authority and responsibility comes from.

It is important to note that this is a media policy choice more than one imposed by the technology, as is illustrated, very presciently, by some 1970s experiments at the BBC. David Attenborough, as Director of Programmes, established the Community Programme Unit in 1973 to facilitate content-creation by “communities”. Anyone could submit proposals to the Unit and it would allocate technical means to the selected projects. There was a weekly BBC2 TV slot late on Monday nights for “Open Door", the umbrella programme that carried the content produced in this way. Writing in 1972, Attenborough saw the value of the concept as being about promulgating “new editorial attitudes that do not derive from the assumptions of the university-educated elite who are commonly believed to dominate television production." He looked forward to success in these terms: “There is a good chance, therefore, that community programmes could broaden the spectrum of views currently broadcast, could occasionally throw up real originality in programme style, and if they are really successful, could produce something that we have tried to achieve many times and signally failed—a genuine correspondence column of the air."

He anticipated the possible dangers of these programmes as follows:

1. they might interfere with the delicate balance achieved by agreement with the political parties;
2. the series might become heavily weighted in one particular political or social direction;
3. they may be dull, boring programmes which are largely unwatched;
4. they might be used as a means of causing sensational outrage for its own sake;
5. they might make the Corporation liable, as publishers, to legal proceedings for libel, contempt of court, and so on.

Attenborough goes on to outline how these risks would be mitigated:

"The first three of these depend on the selection process, and a set of criteria for eligibility will be essential. It is proposed, therefore, that the Programmes Unit should investigate applications and then make formal recommendations to a Selection Committee. This committee should be chaired by D.P.Tel. with E.N.C.A. and C.BBC-2 as permanent members. Their aim should be, within the limits of the editorial code, to provide a lively, varied and entertaining series that was not a corner for cranks nor weighted in one particular direction. The dangers that the programmes might become gratuitously sensational or offensive or infringe the law can be tackled by an editorial code which all contributors would be required to sign."

In the Attenborough memo, we can see pre-figured all the key questions that have arisen for editorial choices for the creators of user-generated content. Should there be pre-screening? How does one stop a space being captured by a particular set of political opinions? How to avoid attractive but incendiary content? Or the dominance by “cranks”? The fifth risk – the financial liability of the publisher for content beyond their control – was famously averted by the platforms through regulatory change: in the US, the famous Section 230 of the 1996 Communications Decency Act (47 U.S.C. § 230) which offered a platform a safe harbor from litigation relating to content posted by a platform user. The EU and UK followed soon after.

The memo imagines a public service social media conceived of as the embryo of a democratic public sphere: it is “not derive[d] from the assumptions of the university-educated elite”, it could become “the correspondence column of the air" and not a corner for cranks nor weighted in one particular direction"). Most intriguingly, the memo shows that this conception of the democratic public sphere is something that is in the BBC's DNA. The memo acknowledges the difficulty of achieving this, but is willing to take risks and experiment. The Community Programme Unit was disbanded in 2002. Of course, there is a sense in which the Internet of 2002 might have looked like one big Community Programme Unit, and was therefore no longer needed inside the BBC. However, the optimism around internet libertarianism would be short-lived. The unit was shut down when its experience was about to be most needed.

The BBC, by deprioritising these questions and experiments just when they would become the central questions of epistemic policy, has handed to the Californian platforms and TikTok the place at the centre of the nation's public sphere. It was forced - often by government regulation – to retreat into a comfortable space of continuing to do well what it has done for generations. However, the abandonment of the public sphere means that it is losing the country. Now, 74% of 16-24 year-olds say that their primary source of news is social media; online platforms are a news source for 70% of the population, more than the 68% who cite TV. When it comes to entertainment, YouTube is quoted as the preferred destination for 4-15 year-olds while under-25s rely "almost entirely" on social media, video on demand, and streamed music. Fewer than half watch live TV monthly.

Of course, the BBC feeds the social-media channels that attract this attention. However, the channels themselves are, for the most part, where the “editorialising" happens. If YouTube's algorithm, having noted your penchant for Adam Curtis (amongst much else), starts to point you down lower quality conspiratorial content, perhaps ending in anti-vax spaces that start to persuade you, that would be a case of BBC content being used with a large externality cost that is outside the control of the content producer.

The BBC has become one content producer amongst many, rather than something quite different: the main orchestrator of the domestic public sphere. This is reflected in the questions that it (together with DCMS, in the consultation on charter renewal) does ask about its relationship to platforms and social media. For example, the consultation asks whether the BBC should make "greater use of third-party platforms" (Question 14) to share content on YouTube and TikTok. Thus the BBC is positioned as a content supplier to existing ecosystems rather than seeking to construct a digital public sphere. The observation that "other broadcasters (like Channel 4) share full-length episodes on YouTube to reach younger audiences" indicates that DCMS sees platform presence as a distribution strategy, not as an assertion of control over the rules and conventions of public speech. The consultation acknowledges the "revolution in the media landscape" where "lines between curated news and user-generated content are becoming dangerously blurred." The implication is that curated news is safe, user-generated content (UGC) is dangerous, without any thought that UGC is an essential part of a democratic public sphere that ought to be curated, as was the philosophy of the Attenborough memo.

There is a fundamental point about how DCMS conceives of the institution: is it the master-creator of a narrative, or is it the orchestrator of interactions out of which a narrative emerges? As Mark Oliver, media strategist and onetime director of strategy at the BBC has said to me in conversation: “In broadcast, the value comes from crafting the narrative – that is, from content; in social media, the value instead is in designing the interaction".

It is perhaps instructive to compare the operation of the BBC to the operation of Parliament a microcosm of a rule-bound public sphere that we understand well. In Parliament, the rules of debate, their enforcement by a Speaker and by procedure are almost entirely separate from the content, which is determined by the choices of MPs and of the government. There is no sense that, in the ordinary course of events, the Speaker is responsible for what MPs say or what governments place on the agenda. Of course there are cases that stretch the strict separation of content (ie, narrative) from rules of engagement; the Brexit debacle revealed them to a fascinating degree. However, even there, it is clear that the maintenance of a public sphere was separate from the weaving of an emergent narrative, something that was controlled by no one body.

The UK has all the assets needed to experiment to create a digital public sphere

The design of the rules of social interactions in cyberspace has been left largely to Californian and Chinese profit-maximising actors, whose commercial incentives have plausibly led to a whole range of harmful social outcomes. There is a great opportunity for the UK to employ some of its unique capabilities to reshape a digital public sphere for public good.

The public sphere, of course, is no longer confined to country borders as, for the most part, it was in the print and broadcast eras. Content in cyberspace knows few technical barriers to international flow, and even linguistic barriers are weakening with improvements in machine translation. However, the notion of “national conversations" has not been made redundant. The country is woven together at many levels – from sport to political institutions to weather to utilities to civil society groups to a history and a future. All of these create their media, and the attention this media invites is much more national than global. Of course, other countries' conversations sometimes find a place within the national conversation – Hollywood has done this for a long time, as do many streaming series today. But even today that is not so dominant as to counter a national taste and a distinct national conversation. The UK therefore, in thinking of shaping a digital public sphere, would think of policy that shapes the country's epistemic cyber-environment, while understanding that this will have impacts beyond the country – impacts that constitute important facets of soft power – and that what other countries do will affect the overall epistemic environment here.

Other aspects of sovereign policy come to mind as examples, for example finance and the environment. In both cases, states are sovereign but the nature of the space created by policy imply substantial interdependencies. It is very hard, for example, to reduce the mobility of capital without international agreement; and it is often impossible to tackle environmental problems without the cooperation of neighbours. Compared to these, policy latitude to shape the epistemic environment is in some senses broadly similar. UK citizens will always be able to “vote with their mouse" and find other content, just as investors in UK gilts can switch to US Treasuries. But the space of what is encouraged or discouraged before that “exit” occurs is large. Moreover, as was seen in the BBC's influence in the broadcast era, a successful pro-social intervention by the BBC will influence taste and expectations and therefore change what it is that the market will offer in terms of alternatives. That is perhaps where the power of “market shaping by constructive intervention" is greatest: it not only offers an alternative, but shapes the offerings in the market through its impact on demand.

There are three essential inputs to the creation of a digital public sphere: attention, algorithmic innovation and digital infrastructure. The UK can bring together resources from the BBC, from innovators in our media industries, digital industries and universities and can partner with some of the non-media digital hyperscalars (eg, Microsoft, or perhaps some new sovereign or European entrants to the space) to rebuild the digital public sphere.

In the media torrent of today, perhaps the capability that is hardest to come by is the attention of large numbers of people. The BBC as a loved and trusted content producer still has the ability and brand to gather large amounts of the attention of the country. A small fraction of that attention should be channeled through an innovation programme towards experiments in designing rules for social connection. The most promising innovations should be scaled. The innovation programme itself should involve input and contributions from the range of the best place institutions in the UK, as was envisaged, with a rather narrower scope, in the Cairncross review. Evaluation should be carried out against strict criteria to determine where to scale and where to park innovation.

In many of the experiments that ought to be considered, a natural question will be "if it's such a great idea, why does the market not offer this already?” There will be many specifics to the answer for specific experiments. But it is worth saying a few things in general about barriers to entry and the nature of market failure in areas characterised by complex network externalities:

• Social media have high fixed costs and low variable costs, making scale a commercial advantage and oligopoly likely
• Add to that the presence of network externalities of the sort characterised by the creator/contributor thinking “I go where the audience is”, and the audience thinking “I go where the creator contributors are”, there are likely to be only a small number of destinations, and therefore a small number of possible products tested and trialled
• Network externalities make it costly to exit a platform, since the value comes in large part from the whole group of users it has attracted
• Cory Doctorow's now famous "logic of enshittification” can be characterised as a demand-side failure: users are attracted to a good feature set at first, but once a platform is established and exit is hard, the quality to them is degraded in favour of returns to shareholders.

Market failures abound in online media, and the market cannot therefore be trusted to generate the kind of media that large numbers of citizens would want to see and, in some form, to fund. In the next two sections, the paper provides some examples of experiments that might be performed in this programme to reshare the country's digital public sphere. These are not meant to be well-formed proposals. They are intended to get others with the right skills to come forward with suggestions.

A few high level examples of experiments that might be considered to get the imagination going

These examples have all been constructed from the premise that existing commercial online media products have virtues which the public-service versions should try to encapsulate without allowing the other features, driven by a singular drive to profit maximisation, to dominate. The examples are labelled by the existing services they are inspired by.

Reddit

At its core, Reddit is a brilliant component of the public sphere. Anyone with an interest-area can set themselves up as the moderator of a community and try to bring a discussion to life. In the language of media strategy, Reddit allows people to create communities of interest around “passion verticals" (eg, hobbies). As anyone who has run a forum or a comments section well-knows, the real difficulty of doing this is the moderation – excluding the contributions from wreckers and from those who have no intent on pursuing a genuine discussion.

Reddit offers community-builders various moderation tools as well as guidance for moderators. However, either through inability or through free-speech libertarianism, Reddit is permissive of a great deal of speech that has no place in a democratic public sphere. Reddit does not see its purpose as building that public sphere, and therefore does not try to move its violent or hateful communities away from their comfort zone. A public service version of Reddit would keep much of the structure that is there. However, it might experiment with certain changes:

• No anonymity, and an enforcement of real identity or perhaps a controlled pseudonymity (eg, ensuring that the moderator can find the real identity while allowing pseudonyms for most viewers); this could include publicly verifiable indicators of where someone is based
• No bots – proof of humanity tools
• Moderation tools allowing the use of cross-community scores of good public-sphere behaviour
• Recommendation engines based on algorithms that include: “most persuasive counter-argument”; “communities that share my interests but not my opinions"
• Awards, recognition – and perhaps even rewards – for good moderation and good participation in the public sphere

X/Bluesky

X (and its "nicer” twin, Bluesky) are also potentially very good components of a well-ordered public sphere. Unlike Reddit, they do not try to separate conversations into communities, and something rather more permeable than Reddit-style conversations is achievable in these media. However, as X has shown, especially after its migration from Twitter, it too can become an unattractive public sphere because of the lack of moderation, or tools to have effective "self-moderation". This is again where algorithmic and tool-based innovation could have a great deal of impact. Parameters that a public-purpose version of these tools could experiment with are:

• Some of the same identity/pseudonymity experiments as for Reddit
• Tools in the hands of individuals to exclude from their feeds the ability to see the contributions from those who fail to satisfy aggregate conditions across the whole network (eg use of violent language, etc)
• Algorithmic choice in recommendation algorithms – allowing people to search for high quality countering views and accounts

TikTok/Instagram/YouTube

Short form video media have proved a remarkable form for the promotion and demonstration of “passion verticals” – sports, hobbies, etc. But of course, some passions should not be encouraged. This is obviously the case with the self-harm content that continues to proliferate on these platforms. Moreover, passion has addictive qualities and these should be managed carefully. A public-service version of these media would recognise their strengths and experiment with ways of reducing their harms. The obvious experiments here should be the algorithmic feed and recommendation engine. However, experimentation with formats that cross-over short-form video with the popular broadcast "passion verticals" like baking, sewing and gardening have clear potential. (A specific idea around the sewing bee is provided below).

ChatGPT, Character.ai, etc

LLMs seem quite likely to generate new media products of their own. Already, it is clear that the creation of synthetic companions can be extremely engaging for some people. Mark Zuckerberg has spoken of his desire to supply us with the friends that surveys reveal we wished we had. Chatbots are an extremely magnetic medium for some people, and cases where this has become problematic have been much publicised. This is one of the media where experimentation and innovation are important and at an early stage. A version of Character.ai where companions are not essentially pornographic, and where conversations have some degree of automated oversight would certainly be one that many parents would prefer to give access to their children. Similarly, chatbots with automated oversight to try to reduce the risks of “Ai induced psychosis" would be attractive to many people.

Data collection and use

Attention harvesting has generalised surveillance for two reasons: first, as input to feed-creating – the algorithm needs to predict the behaviour of individual faced with different pieces of content to keep them scrolling – and second for advertising: the advertiser wants to target particular types of people, and the more information the platform can provide, the greater its attractiveness to paying advertisers. This twin use of surveillance has given data collection something of a bad name.

However, data collection can obviously be imagined for good – and even when it comes to advertising. If data harvesting were used for the good of the person and their community first and foremost, then data collection and use could be transformed into a positive feature of media viewing habits. For example, many people are interested in understanding themselves better – the popularity of psychological quizzes online are a testament to this – and tools that use data to help and advise on health, wellbeing, companionship and community are a part of a good public sphere. Not, however, when the advice is really a front for profitable data collection. A public sphere organised for public good would experiment with data collection and tracking mechanisms that were both ethically acceptable and useful.

Three more specific ideas for product/programme trials

If the above was meant to give examples of “where existing social media could be given a public-service scrubbing", the next three examples are ones where the approach is rather to say: "where could existing BBC products be given some social media shaping attributes".

Pilot 1: BBC Match Day

Imagine a Saturday evening where Match of the Day doesn't end when the credits roll. Alan Shearer, having just dissected Liverpool's high line, turns to camera: "There's a fascinating thread developing on BBC MatchDay right now about that offside trap—over 200 tactical analysts are debating whether it's sustainable against City next week." The audience, instead of turning to the toxic swamp of Football Twitter, finds themselves in a space where the quality of analysis matters more than the volume of outrage.

This would be a dedicated community for football fans built on Reddit-like foundations but with a crucial difference: reputation would be earned through accuracy and insight, not follower counts. Users would accumulate badges, "tactical analyst," "club historian" – based on the quality of their contributions and the accuracy of their predictions. The loudest voice would no longer win; the most thoughtful would. Super-fans, vetted and trained, would moderate their communities with BBC guidelines providing the framework, ensuring that the passion of football debate isn't confused with personal abuse.

The platform would launch through partnership with a data-forward club like Brentford or Brighton, whose analytically-minded fanbase would seed the community with quality discussion. Pundits wouldn't just broadcast into the void but would actively participate, hosting "ask me anything" sessions with users who've earned high reputation scores. The trial would run for a full Premier League season or through a major tournament like the Euros, with success measured not merely by user numbers but by the quality of discourse – tracking the ratio of constructive debate to abusive content compared to existing platforms.

Pilot 2: The Workshop

The Great British Sewing Bee ends with Patrick Grant admiring a contestant's French seams. But instead of viewers simply switching off, inspired but isolated, they hear: "download tonight's pattern from The Workshop and upload your attempt – Patrick will be reviewing submissions this week, and the best makers will receive our digital Blue Peter badge." This is the vision for The Workshop: a short-form video platform dedicated entirely to making, cooking, sewing, gardening, DIY. It captures the creative energy of TikTok while stripping away the misinformation and toxicity.

The platform would operate on a principle that turns the OnlyFans model on its head: creators would build reputation and even earn money through helpfulness rather than exposure. Unlike TikTok's wild west of dubious life hacks, The Workshop would partner with universities and professional guilds to verify content. When someone posts a tutorial on rewiring a plug or fermenting sourdough, viewers would know it's been checked by people who have a reputation for knowing what they're talking about.

The algorithm would be deliberately tuned away from engagement-at-all-costs toward skill-appropriate content. A novice baker would see basic bread recipes, not intimidating croissant techniques designed to make them feel inadequate. Political discussion would be explicitly excluded. This is a space for making, not arguing. Users could tip creators from a communal pot, rewarding genuinely useful content. Universities and colleges would contribute high-quality 60-second explainers that would populate feeds between user submissions, raising the overall quality of the platform, and their contributions would count towards their impact metrics.

The trial would launch alongside a specific programme run, perhaps the ten weeks of The Great British Sewing Bee, with Jay Blades or Patrick Grant serving as figureheads who actively comment on user submissions. Success would be measured not just by how many TV viewers create profiles, but by how many actually complete projects they've started, building a community of makers rather than merely watchers.

Pilot 3: BBC Common Ground

Question Time has always been democracy as theatre, a carefully managed collision of views in a village hall. But what if that model could be evolved for the digital age, creating a space where the entire nation could participate in civic debate without it descending into a shouting match? BBC Common Ground would be that evolution: a news platform where the rules of engagement are strictly enforced to promote understanding over polarisation.

The platform would require real identity verification, perhaps through bank ID or a partnership with Microsoft's LinkedIn, changing the dynamic from anonymous trolling to accountable discourse. Drawing on the best innovation from the worst platform, it would adopt X's Community Notes feature but with BBC editorial oversight. When someone posts a deceptively edited video of a politician, the community could attach context, but unlike X's free-for-all, these notes would be prioritised according to BBC-wide reputation scores, and some number would even go through actual BBC fact-checking before being stamped as authoritative.

Most radically, the recommendation algorithm would be explicitly coded to bridge divides rather than deepen them. Instead of serving up content that confirms biases, it would surface 'the best argument from the other side' to ensure users understand what their fellow citizens actually believe rather than caricatures of those beliefs. The platform would integrate directly with the BBC News app, replacing the current "Have Your Say" comment sections, which too often resemble a digital bear pit, with this more structured environment.

Digital town halls would follow Question Time broadcasts, where the online audience could participate, but only those who have demonstrated consistent civility would have their questions considered for the following week's programme. The trial would run for three months outside any election period to avoid partisan capture, with success measured by how often people with opposing political views manage to interact constructively. The key metric wouldn't be agreement, democracy doesn't require that, but whether the platform creates a space where disagreement can be civil and productive, and whether users whose contributions are initially rejected return with modified approaches rather than storming off to darker corners of the internet.

Subscriptions, license fees and advertising

Public spheres – epistemic environments – do not come for free. That is what this paper started with – the obvious point that what we see is what gets funded. The paper has argued that funding via profit-maximising firms is no way to build a good epistemic environment. However, what can we say about how a public sphere ought to be funded, and what of the question of how to fund the specific innovation programme argued for above?

Media – that which is presented to our attention – is a complex mixture of private good and knock-effect. In the language of economics, it is full of externalities. Think even of what looks most of all like a private good – say a novel. I buy the latest Ian McEwan, I read it, and then I have a conversation with someone about it. That person then buys it and reads it, and, say, writes an essay about it that encourages others to think differently about the climate crisis. In the standard economic story of private goods, I read McEwan because I expect it to generate more value for me than its cost (including, of course, the opportunity cost of the time spent reading it). And yet, in my calculus, there is not much room for the knock-on effects of my reading and the possible pluses and minuses of that. A bottom-up, economist's answer to the question of what media should be produced, seen and talked about seems like an impossibility. And that is not just because of the sorts of externalities described with my reading of the McEwan. It is even more fundamental: what we attend to defines who we are, what we want and how we behave towards each other. Media makes culture, and culture, to a great degree, makes society.

This is all to say that there is no simple laissez-faire argument to say that market incentives alone will create the epistemic environment that we want to live in. There is therefore no simplistic libertarianism that determines the right payment model for media. Having got that out of the way, it is worth pointing out some simple aspects of the economics of the three popular mechanisms for funding epistemic environments: advertising, license fees and subscriptions.

Under the advertising model, funds flow from consumers – those who ultimately pay for the goods, part of whose costs include the advertisement of those goods – to the advertising industry, which takes a cut of those funds for matching commercial messages to content, to content creators, who produce the attention-harvesting material that attracts the audience. It is worth noting that in most advertising/media relations, there is no strict correspondence between those who pay – say the people buying a specific brand of soap – and those who enjoy the content that the advertising paid for – say a specific soap opera.

Of course, the advertisers hope to convert the audience into buyers of that soap, and thus create a link, but there is nothing automatic about it. Many people will buy the soap, and thus pay for the soap opera, without ever having seen or had any interest in the soap opera. There is a real sense, therefore, in which the advertising model is one that socialises the cost of media. Disregarding for a moment the economically useful aspects of some advertising – like informing the consumer – one could even say that funding media through advertising is financially identical to funding it through a sales tax, especially given that the goods that attract the most online advertising spending, like clothing, electronics, financial products, package holidays and housing are essentially not available in unbranded forms that avoid the "advertising tax” . If, say, 1% of the purchase price of all consumer goods went to fund the same content as was funded by the advertising industry, there would be no difference to consumers or to content makers.

In this way, funding by advertising is perhaps not entirely different from funding by license fee. Of course, it is an exaggeration to say this – this entire paper has been premised on the way that advertising has influenced the nature of the digital public sphere. However, this point about flow-of-funds is important when thinking about objections to public funding mechanisms like a license fee, an advertising levy or a tax. All these mechanisms create content that can be free-at-the-point-of-use, and all of them take funds out of consumers' pockets. Indeed, the free-at-the-point-of-use web costs every UK household about £1,000 per year, about six times more than the BBC license fee. Moreover, that £1,000 per year is not practically avoidable in any way, much as a license fee has become hardly avoidable. Of course, as the story of US radio showed, advertising is a sort of "populist vote" mechanism for determining content, while the license fee, as we have discussed, sets up a different set of institutional and democratic incentives.

There are some interesting “popular vote" funding mechanisms that have developed in cyberspace that are not based on advertising. Patreon is the best known, with direct and voluntary payment opportunities as "thanking" the amateur creator. These are similar to the methods that Father Coughlin used when he was eventually excluded from the ad-funded national networks: he solicited direct donations to his church. There are some interesting experiments that the BBC could try with these funding mechanisms. For example, audiences could have a budget of credits that they could allocate to amateur creators in a Patreon-like way, with some level of credit funded by money gathered from levies and taxes.

As against advertising and the license fee, subscription models really are different from an economic perspective, especially the pure-form ones that do not complement a subscription with advertising – for example some of the Netflix or Spotify plans. Under these models, the person does from time to time make a decision about whether to fund the service they are receiving: my Netflix subscription is avoidable to me in a way that my advertising-fee to the open web is not. We can certainly imagine a digital public sphere in which only pure subscriptions were allowed, and they might have certain attractive properties. However, we should note that no society has ever existed in which the free-at-the-point-of-use epistemic environment is only subscription-based. Humans get together and create culture, one way or another. The attention-space is one that gets filled by one process or another. The argument often advanced by the Murdoch empire that the BBC should compete on the basis of subscriptions like its satellite channels do is really a denial that the epistemic environment should be shaped by collective decision-making.

Enough theory and throat-clearing. Does the proposal that the BBC, in collaboration with other key institutions, should take an active role in constructing our national epistemic environment imply anything about how it should be funded?

In general, funding models for complex public goods are a matter of judgement. They often include a mix of user charges, taxes, commercial revenues and more. In this mix, there is a great deal of experimentation to be carried out with regard to funding as well as with regard to the shape of the public sphere itself. For example, it is clear that attention is a very valuable asset and is one that the BBC does not directly monetise. However, there are ethical ways to monetise attention. For example, a viewer watching Martin Lewis talk about energy prices might decide that they were going to explore energy tariff options. At that very moment, they might have some way of announcing: “I have the next 10 minutes to make this decision. Convince me, perhaps by means of payment, that I should spend any of that 10 minutes looking at your offer”. In other words, there is experimentation to be done whereby citizens, when they choose to act as consumers, auction their attention directly rather than handing that over to intermediaries like Google or Meta. The organisation of such attention-auctions could be done by the orchestrator of the public sphere, and those earnings to consumers could be offset against payments to the orchestrator. If the BBC cost you £175 per year, but through its attention-orchestration allowed you to earn the £500 per year that Google and Meta are currently earning from it, perhaps the politics of the license fee would become less divisive.

Conclusion

The BBC stands at a crossroads. It can continue its defensive crouch, producing excellent content for platforms that increasingly determine who sees it and how. Or it can reclaim its historic role as the architect of Britain's public sphere, extending into cyberspace the same civilising mission that transformed radio and television from potential instruments of demagoguery into pillars of democratic society.

This is not a task the BBC can or should undertake alone. A partnership between the BBC's R&D department with institutions like the Turing Institute, Nesta and Open Data Institute - potentially supported by hyperscalers like Microsoft – should launch an ambitious programme of experimentation. These experiments would test new models for a democratically controlled digital public realm, exploring everything from novel funding mechanisms to algorithmic designs that promote understanding rather than outrage.

The path forward requires DCMS and the BBC to convene a working group that can rapidly move from concept to trial, selecting the most promising proposals for development. Crucially, government must also examine and repeal the inherited restrictions on BBC innovation that were products of a different era's regulatory fashion—constraints that now prevent the Corporation from building the digital infrastructure our democracy needs.

The window for action is narrowing. Every month that passes sees the BBC's influence over national attention diminish, while the pathologies of our platformised public sphere deepen. But the opportunity remains extraordinary: to demonstrate that a democratic society can still shape its own information environment, that public service can compete with private extraction, and that Britain can offer the world something better than the current choice between Californian surveillance capitalism and Chinese digital authoritarianism. The BBC was created to civilise broadcasting. Its next century depends on whether it can civilise cyberspace.

Annex 1 - Five arguments against the proposal, and a response

Financial viability: Running such a platform is extremely difficult given cybersecurity and operational challenges. The BBC cannot do this.

This misunderstands the suggestion and also ignores the UK's unique sovereign capabilities. We are not proposing that the BBC builds a monolithic rival to Meta from scratch using internal IT procurement. Instead, this proposal envisages a venture-studio model, orchestrating the best of the UK's world-class digital sector, university talent, and cybersecurity expertise (such as the NCSC). We actually think that scaling should be done with a hyperscalar – perhaps Microsoft would be a natural partner as the one not primarily in the attention-harvesting market.

The "crack" problem: Social media's addictive nature is its strength – a "virtuous" alternative may not attract users

There is a cynical view that the public only wants "digital crack"—that without the dopamine loops of outrage and addiction, a platform will be a ghost town. This ignores the BBC's founding purpose: "to inform, educate, and entertain." A public service algorithm need not be a vegetable patch of dull worthiness; it can be designed for genuine engagement rather than compulsive extraction. We see in the rise of long-form podcasts, "slow media," and niche community discords that there is a massive, unsatisfied appetite for connection that is nourishing rather than depleting. The market failure is not that people don't want healthy digital spaces; it is that the current advertising model cannot monetise them as ruthlessly as it does addiction.

Network effects: How do you compete with established platforms that already have massive user bases?

Critics rightly point out that you cannot simply wish a network effect into existence when incumbents have billions of users. However, the BBC possesses a "super-weapon" that no startup has: the continuing ability to direct the attention of much of the nation. Through its broadcast channels, iPlayer, and Sounds, the BBC commands millions of hours of daily attention which can be used to solve the "cold start" problem, driving audiences toward these new experiments. Even in the youth segment, the BBC still enjoys substantial attention, and the prospect of worried parents allowing children to use and view BBC-partnered cyber-properties means it still has the power to shape norms of online behaviour for the future. Moreover, the incumbents are currently vulnerable; as platforms undergo "enshittification"—degrading user experience to squeeze out profit—users are actively looking for lifeboats. Of course, that power is diminishing and it is urgent to start the work of converting remaining influence into an enduring power.

BBC brand limitations: Are we overestimating the BBC's brand penetration, especially among younger audiences?

It is true that the BBC brand carries baggage, particularly with younger demographics who associate it with linear television or establishment views. However, this proposal is not about slapping the BBC logo on a social network; it is about the BBC providing the infrastructure of trust. Young audiences are actually the most discerning critics of authenticity and the most exhausted by the fake reality of the influencer economy. A platform that offers transparency, data sovereignty, and freedom from commercial surveillance appeals directly to values held strongly by digital natives. The BBC's role is to be the guarantor of the protocol, not necessarily the face of every interaction.

Risk of creating "new legacy media": With AI/LLMs evolving (like Character AI), there's a risk of building something that becomes obsolete

There is a danger that in trying to fix social media, we build a solution for 2015 just as the world moves to 2030's AI-driven reality. This is exactly why we propose a Nesta-led innovation lab rather than a monolithic procurement project. A rigid "BBC Facebook" would indeed be obsolete on arrival. A venture studio, however, is designed to experiment with the frontier of interaction, including the LLMs and synthetic companions mentioned earlier in this paper. The goal is not to clone Reddit or X, but to apply Reithian principles to whatever the dominant medium becomes—whether that is a text feed today or an AI agent tomorrow. By stepping into the arena, the BBC ensures it is a player in the next technological wave, rather than a victim of it.