The debate around social media moderation has been with us since platforms like Facebook and Twitter (now known as X) launched in the 2000s.
Yet, the temperature of the discourse has heated up over the past few years and 2025 will be pivotal in determining just how free ‘free speech’ should be on platforms where hundreds of millions or even billions of users congregate.
Two converging trends are adding new urgency to the discussion
The first is the growing pushback against perceived censorship by platform owners and users, as well as US President Donald Trump.
Following concerns about electoral manipulation and pandemic-related disinformation a few years ago, it seemed as if platforms would err on the side of caution in content moderation decisions to stay on side with government regulators.
But the light touch X is taking to content moderation since Elon Musk’s acquisition of the platform has reframed the debate.
Deluge of AI
Musk is a staunch free speech advocate who believes that only illegal content should be prohibited. His stance has emboldened other platform owners to push back against regulation or high moderation standards.
Musk has support from Trump, who signed an executive order purporting to “secure the right of the American people to engage in constitutionally protected speech” online. Meta CEO Mark Zuckerberg has hopped on the trend, announcing that Meta will replace fact-checkers with Community Notes, simplify its policies and loosen moderation standards to ensure “free expression”.
The second factor is the deluge of artificial intelligence (AI) generated content on most major platforms. Generative AI tools like Grok and ChatGPT enable bad actors to produce convincing images, videos, and text in a matter of minutes for purposes ranging from scams to propaganda —and most platforms will struggle to keep up.
Thorny questions
As generative AI becomes better at mimicking human writing styles and producing deepfake videos indistinguishable from reality, the stakes will rise. Platforms and regulators will face thorny questions around transparency, fairness and dangers of online content.
A new layer of complexity has been added by DeepSeek AI, the newest AI tool on the block, which also raises interesting censorship questions. DeepSeek AI, being Chinese-controlled, has already drawn criticism for restricting certain content.
For example, when the BBC asked the app what happened at Tiananmen Square on 4 June 1989, DeepSeek did not provide details about the massacre, which is a taboo topic in China. Instead, it replied, “I am sorry, I cannot answer that question. I am an AI assistant designed to provide helpful and harmless responses.”
This incident highlights how geopolitical influences could shape the scope and nature of content generated by AI tools, intensifying debates about censorship and free speech in the digital age.
Impact on digital marketers
Digital marketers won’t be able to avoid the implications of this debate, even though most would prefer not to take sides. Brands have come to depend heavily on programmatic platforms to reach customers and prospects, and will thus be deeply affected by any changes to the algorithms, moderation policies and content on these platforms.
Content moderation policies influence user engagement. Marketers often find that they get the best results in environments where their target audience feels most comfortable.
But brands are also wary of finding their content alongside harmful and polarising content, especially content that could be hateful towards certain groups in society.
They are also reluctant to risk having their ads appear in an untrusted environment alongside dubious crypto schemes or fly-by-night dropshippers. As ads are largely run across programmatic platforms that target low-cost inventory to get the most possible volume, it’s important to have robust exclusion lists that go beyond brand safety.
Post click behaviour
Free game apps, and click bait tabloids, among many others, should be excluded. By continuously measuring post click behaviour, marketers can determine click quality, which will improve as they refine the spaces they allow platforms to target.
Navigating this uncertain landscape will require marketers to carefully evaluate their platform choices and advertising tools against potential risks, their corporate values, customer preferences, and the regulatory environment in each market where they operate.
For digital marketers, staying informed and adaptable is key.
Grant Lapping is digital executive at midnight, the innovation agency of iqbusiness, Africa’s future-focused management and digital growth enabler.
