Content Policy

Hundreds lose jobs as TikTok focuses on AI moderation

TikTok, owned by ByteDance, is cutting hundreds of jobs globally as it pivots towards greater use of AI in content moderation. Among the hardest hit is Malaysia, where fewer than 500 employees were affected, mostly involved in moderation roles. The layoffs come as TikTok seeks to improve the efficiency of its moderation system, relying more heavily on automated detection technologies.

The firm’s spokesperson explained that the move is part of a broader plan to optimise its global content moderation model, aiming for more streamlined operations. TikTok has announced plans to invest $2 billion in global trust and safety measures, with 80% of harmful content already being removed by AI.

The layoffs in Malaysia follow increased regulatory pressure on technology companies operating in the region. Malaysia’s government recently urged social media platforms, including TikTok, to enhance their monitoring systems and apply for operating licences to combat rising cybercrime.

ByteDance, which employs over 110,000 people worldwide, is expected to continue restructuring next month as it consolidates some of its regional operations. These changes highlight the company’s ongoing shift towards automation in its content management strategy.

Cybercriminals use AI to target elections, says OpenAI

OpenAI reports cybercriminals are increasingly using its AI models to generate fake content aimed at influencing elections. The startup has neutralised over 20 attempts this year, including accounts producing articles on the US elections. Several accounts from Rwanda were banned in July for similar activities related to elections in that country.

The company confirmed that none of these attempts succeeded in generating viral engagement or reaching sustainable audiences. However, the use of AI in election interference remains a growing concern, especially as the US approaches its presidential elections. The US Department of Homeland Security also warns of foreign nations attempting to spread misinformation using AI tools.

As OpenAI strengthens its global position, the rise in election manipulation efforts underscores the critical need for heightened vigilance. The company recently completed a $6.6 billion funding round, further securing its status as one of the most valuable private firms.

ChatGPT continues to see rapid growth, boasting 250 million weekly active users since launching in November 2022, emphasising the platform’s widespread influence.

Temu faces deadline from EU over illegal product sales

The European Commission has set a deadline of October 21 for the Chinese online marketplace Temu to respond to inquiries regarding its compliance with the Digital Services Act (DSA). The Commission is seeking detailed information about Temu’s efforts to combat the sale of illegal products on its platform and the measures it has implemented to ensure consumer protection, public health, and user wellbeing.

Temu, founded in 2022 by PDD Holdings, was classified as a Very Large Online Platform due to its user base exceeding 45 million monthly average users in the EU. It was previously required to meet DSA standards by the end of September, including addressing systemic risks and preventing the sale of counterfeit goods. This latest inquiry marks the second time the Commission has sought clarification from Temu, following questions in June about its compliance with the “Notice and Action mechanism” for reporting illegal products.

The European Consumer Organisation (BEUC) has also raised concerns about Temu’s practices, filing complaints against the platform for failing to protect consumers and employing manipulative tactics. These complaints, supported by representatives from 17 EU member states, allege that Temu does not provide essential seller information, hindering consumers’ ability to verify product safety compliance. The DSA has been in effect since February, and the EU has initiated several investigations into other major platforms for similar compliance issues.

X returns to Brazil as court clears path for resumption

Social media giant X, formerly known as Twitter, became accessible to some Brazilian users on Wednesday, just one day after the country’s Supreme Court cleared the platform to resume operations by complying with court rulings. Brazil’s telecommunications regulator, Anatel, announced that it had begun instructing internet providers to restore access to X. Many users celebrated the return of the platform, with topics like ‘we’re back’ trending across Latin America’s largest country.

Despite the reopening, some Brazilians still encountered difficulties accessing X, as Anatel indicated that the restoration time would depend on the procedures of individual internet providers. Supreme Court Justice Alexandre de Moraes, who had been engaged in a lengthy dispute with billionaire Elon Musk, granted approval for X’s return on Tuesday afternoon. He instructed Anatel to ensure the platform was operational within 24 hours, affirming that X had fulfilled all necessary requirements to resume its services.

X had been suspended in Brazil since late August due to its failure to comply with court orders related to hate speech moderation and the absence of a designated legal representative in the country, as mandated by law. As the platform’s sixth-largest market worldwide, Brazil accounted for approximately 21.5 million users as of April, making the resumption of service a crucial step for X’s growth and presence in the region.

Russia blocks Discord over content violations

Russia‘s communications regulator, Roskomnadzor, has blocked the messaging platform Discord for alleged violations of Russian law, according to the TASS news agency. The San Francisco-based company becomes the latest foreign tech platform to face restrictions in Russia. Discord has yet to respond to the decision.

For years, Russia has pressured foreign tech companies to remove content it deems illegal, imposing frequent, though generally small, fines for non-compliance. Last week, Roskomnadzor ordered Discord to delete nearly 1,000 pieces of content it classified as illegal and had previously fined the platform for failing to remove banned material.

Moscow has also blocked other major platforms, including Twitter (now X), Facebook, and Instagram, shortly after the invasion of Ukraine in February 2022.

TikTok faces legal challenges from 13 US states over youth safety concerns

TikTok is facing multiple lawsuits from 13 US states and the District of Columbia, accusing the platform of harming and failing to protect young users. The lawsuits, filed in New York, California, and other states, allege that TikTok uses intentionally addictive software to maximise user engagement and profits, particularly targeting children who lack the ability to set healthy boundaries around screen time.

California Attorney General Rob Bonta condemned TikTok for fostering social media addiction to boost corporate profits, while New York Attorney General Letitia James connected the platform to mental health issues among young users. Washington D.C. Attorney General Brian Schwalb further accused TikTok of operating an unlicensed money transmission service through its live streaming and virtual currency features and claimed that the platform enables the sexual exploitation of minors.

TikTok, in response, denied the allegations and expressed disappointment in the legal action taken, arguing that the states should collaborate on solutions instead. The company pointed to safety measures, such as screen time limits and privacy settings for users under 16.

These lawsuits are part of a broader set of legal challenges TikTok is facing, including a prior lawsuit from the U.S. Justice Department over children’s privacy violations. The company is also dealing with efforts to ban the app in the US due to concerns about its Chinese ownership.

Brazil lifts ban on Elon Musk’s X platform

Brazil’s Supreme Court has lifted the suspension of Elon Musk’s social media platform, X, allowing it to resume operations in the country after the company finally complied with local court rulings. The resolution marks the end of a prolonged dispute between Musk and Supreme Court Justice Alexandre de Moraes, who had previously blocked the platform for refusing to follow legal orders. In his ruling, Moraes stated that X had met the conditions to return online, paving the way for its swift restoration.

The conflict began when Musk, a vocal advocate of free speech, resisted Brazilian court orders to block accounts flagged for spreading misinformation, calling the directives censorship and labelling Moraes a ‘dictator.’ However, in recent weeks, Musk’s platform reversed course, appointing a local representative, paying outstanding fines, and complying with the court’s requests to block certain accounts. By doing so, X earned back its legal right to operate in Brazil.

Brazilian users could not access X on Tuesday evening despite the platform’s return, even though the country’s telecommunications regulator, Anatel, has been instructed to restore the service within 24 hours. Through its Global Government Affairs account, X expressed pride in returning to the Brazilian market, emphasising its commitment to upholding free speech within legal boundaries. Brazil remains X’s sixth-largest market globally, with about 21.5 million users as of April, according to Statista.

Why does it matter?

The dispute between Musk and the Brazilian government is part of a broader struggle Musk has faced with international authorities seeking to regulate online platforms. Brazil’s communication minister, Juscelino Filho, hailed the decision as a victory, stressing that all companies operating in the country must respect its laws regardless of size or influence. President Luiz Inácio Lula da Silva echoed this sentiment, remarking that the world should not have to endure Musk’s ideology simply because of his wealth.

Many users migrated to rival platforms like Bluesky and Meta’s Threads during the suspension, especially with Brazil’s municipal elections underway.

While X remained offline for the election’s first round, the platform could be reinstated just in time for the run-offs, set to take place in late October, including in São Paulo, Latin America’s largest city.

Independent body in Ireland empowers EU social media users to challenge content moderation decisions

A new independent body in Ireland will allow social media users in the European Union to challenge content moderation decisions made by platforms like Facebook, TikTok, and YouTube. Established under the EU Digital Services Act (DSA), this Appeals Centre aims to provide users with an alternative to the courts when disputing content decisions. Supported by Meta’s Oversight Board Trust and certified by Ireland’s media regulator, the centre is expected to begin operations by the end of the year. It will expand to include more platforms over time.

Thomas Hughes, CEO of the Appeals Centre, emphasised the body’s independence from governments and companies, ensuring that social media content policies are applied fairly. The centre’s team of experts will review cases within 90 days to determine if the platforms’ actions align with their stated policies. The European Commission has expressed support for the initiative, with spokesperson Thomas Regnier highlighting the importance of uniform development across the EU to strengthen online user rights.

Located in Dublin, the Appeals Centre will operate on a funding model that charges social media companies fees for each case. At the same time, users will incur a nominal fee that is refundable if their appeal is successful. However, platforms are not obligated to participate, as the centre lacks the power to enforce binding settlements. The centre will be governed by a board of seven non-executive directors.

Study shows ChatGPT tends to overprescribe in emergencies

A study conducted by UC San Francisco found that ChatGPT, when applied in emergency care, often recommends unnecessary treatments such as X-rays and antibiotics. It also admits patients who don’t require hospitalisation. Despite its strengths in certain areas, the AI model struggles to match the accuracy of a human doctor in more complex decision-making.

Researchers discovered that while ChatGPT can excel in simpler tasks like determining which patient is sicker, it tends to overprescribe when faced with real emergency cases. ChatGPT-4 performed 8% worse than resident doctors, while version 3.5 was 24% less accurate. These overprescriptions could lead to unnecessary treatments, increased healthcare costs, and strain on resources.

The research highlighted that the AI models are influenced by their internet training, often erring on the side of caution by recommending medical consultations. Although this approach is appropriate for general safety, it can be problematic in emergency settings, where unneeded treatments can harm patients. More refined frameworks are needed before AI can reliably assist in EDs.

Researchers are working on better ways for AI to evaluate clinical information in emergency care. A balance must be found between preventing serious oversight and avoiding excessive medical interventions.

Independent candidate uses AI to spark debate

In Virginia’s 8th congressional district, independent challenger Bentley Hensel has come up with an unconventional approach to engage incumbent Democrat Don Beyer, who secured a decisive win in 2022. Upset by Beyer’s refusal to join further debates, Hensel developed an AI chatbot called ‘DonBot,’ which mimics Beyer’s public positions. The AI uses OpenAI’s technology and pulls information from Beyer’s official platforms to generate responses. Set for 17 October, the debate will have Hensel and independent candidate David Kennedy facing off against DonBot, with Republican candidate Jerry Torres potentially joining if he responds in time.

Beyer’s campaign stated that he would not attend the upcoming debate, noting that he had already participated in a previous forum and had no plans for additional events. His spokesperson highlighted Beyer’s dedication to regulating AI to curb election misinformation but did not comment on the legality of Hensel’s AI tactic. Legal experts indicate that Beyer has few options to prevent the debate as long as the AI’s use remains non-commercial and clearly identified as artificial.

Hensel, who has raised about $17,000 for his campaign compared to Beyer’s $1.5 million, admits the AI debate is a last-ditch effort to gain attention in a race he acknowledges is heavily tilted in Beyer’s favour. He believes the event will highlight the potential of AI in politics while advocating for more transparency in election campaigns.