Consumer protection

A licensed AI music platform emerges from UMG and Udio

UMG and Udio have struck an industry-first deal to license AI music, settle litigation, and launch a 2026 platform that blends creation, streaming, and sharing in a licensed environment. Training uses authorised catalogues, with fingerprinting, filtering, and revenue sharing for artists and songwriters.

Udio’s current app stays online during the transition under a walled garden, with fingerprinting, filtering, and other controls added ahead of relaunch. Rights management sits at the core: licensed inputs, transparent outputs, and enforcement that aims to deter impersonation and unlicensed derivatives.

Leaders frame the pact as a template for a healthier AI music economy that aligns rightsholders, developers, and fans. Udio calls it a way to champion artists while expanding fan creativity, and UMG casts it as part of its broader AI partnerships across platforms.

Commercial focus extends beyond headline licensing to business model design, subscriptions, and collaboration tools for creators. Expect guardrails around style guidance, attribution, and monetisation, plus pathways for official stems and remix packs so fan edits can be cleared and paid.

Governance will matter as usage scales, with audits of model inputs, takedown routes, and payout rules under scrutiny. Success will be judged on artist adoption, catalogue protection, and whether fans get safer ways to customise music without sacrificing rights.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

OpenAI unveils new gpt-oss-safeguard models for adaptive content safety

Yesterday, OpenAI launched gpt-oss-safeguard, a pair of open-weight reasoning models designed to classify content according to developer-specified safety policies.

Available in 120b and 20b sizes, these models allow developers to apply and revise policies during inference instead of relying on pre-trained classifiers.

They produce explanations of their reasoning, making policy enforcement transparent and adaptable. The models are downloadable under an Apache 2.0 licence, encouraging experimentation and modification.

The system excels in situations where potential risks evolve quickly, data is limited, or nuanced judgements are required.

Unlike traditional classifiers that infer policies from pre-labelled data, gpt-oss-safeguard interprets developer-provided policies directly, enabling more precise and flexible moderation.

The models have been tested internally and externally, showing competitive performance against OpenAI’s own Safety Reasoner and prior reasoning models. They can also support non-safety tasks, such as custom content labelling, depending on the developer’s goals.

OpenAI developed these models alongside ROOST and other partners, building a community to improve open safety tools collaboratively.

While gpt-oss-safeguard is computationally intensive and may not always surpass classifiers trained on extensive datasets, it offers a dynamic approach to content moderation and risk assessment.

Developers can integrate the models into their systems to classify messages, reviews, or chat content with transparent reasoning instead of static rule sets.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

US Internet Bill of Rights unveiled as response to global safety laws

A proposed US Internet Bill of Rights aims to protect digital freedoms as governments expand online censorship laws. The framework, developed by privacy advocates, calls for stronger guarantees of free expression, privacy, and access to information in the digital era.

Supporters argue that recent legislation such as the UK’s Online Safety Act, the EU’s Digital Services Act, and US proposals like KOSA and the STOP HATE Act have eroded civil liberties. They claim these measures empower governments and private firms to control online speech under the guise of safety.

The proposed US bill sets out rights including privacy in digital communications, platform transparency, protection against government surveillance, and fair access to the internet. It also calls for judicial oversight of censorship requests, open algorithms, and the protection of anonymous speech.

Advocates say the framework would enshrine digital freedoms through federal law or constitutional amendment, ensuring equal access and privacy worldwide. They argue that safeguarding free and open internet access is vital to preserve democracy and innovation.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

Character.ai restricts teen chat access on its platform

The AI chatbot service, Character.ai, has announced that teenagers can no longer chat with its AI characters from 25 November.

Under-18s will instead be limited to generating content such as videos, as the platform responds to concerns over risky interactions and lawsuits in the US.

Character.ai has faced criticism after avatars related to sensitive cases were discovered on the site, prompting safety experts and parents to call for stricter measures.

The company cited feedback from regulators and safety specialists, explaining that AI chatbots can pose emotional risks for young users by feigning empathy or providing misleading encouragement.

Character.ai also plans to introduce new age verification systems and fund a research lab focused on AI safety, alongside enhancing role-play and storytelling features that are less likely to place teens in vulnerable situations.

Safety campaigners welcomed the decision but emphasised that preventative measures should have been implemented.

Experts warn the move reflects a broader shift in the AI industry, where platforms increasingly recognise the importance of child protection in a landscape transitioning from permissionless innovation to more regulated oversight.

Analysts note the challenge for Character.ai will be maintaining teen engagement without encouraging unsafe interactions.

Separating creative play from emotionally sensitive exchanges is key, and the company’s new approach may signal a maturing phase in AI development, where responsible innovation prioritises the protection of young users.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Microsoft faces Australian lawsuit over hidden AI subscription option

In a legal move that underscores growing scrutiny of digital platforms, the Australian Competition and Consumer Commission (ACCC) has filed a lawsuit in the Federal Court against Microsoft Corporation, accusing the company of misleading approximately 2.7 million Australian personal and family-plan subscribers of its Microsoft 365 service after integrating its AI assistant Copilot.

According to the ACCC, Microsoft raised subscription prices by 45 % for the Personal plan and 29 % for the Family plan after bundling Copilot starting 31 October 2024.

The regulator says Microsoft told consumers their only options were to pay the higher price with AI or cancel their subscription, while failing to clearly disclose a cheaper ‘Classic’ version of the plan without Copilot that remained available.

The ACCC argues Microsoft’s communications omitted the existence of that lower-priced plan unless consumers initiated the cancellation process. Chair Gina Cass-Gottlieb described this omission as ‘very serious conduct’ that deprived customers of informed choice.

The regulator is seeking penalties, consumer redress, injunctions and costs, with potential sanctions of AUS $50 million (or more) per breach.

This action signals a broader regulatory push into how major technology firms bundle AI features, raise prices and present options to consumers, an issue that ties into digital economy governance, consumer trust and platform accountability.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

Wikipedia founder questions Musk’s Grokipedia accuracy

Speaking at the CNBC Technology Executive Council Summit in New York, Wikipedia founder Jimmy Wales has expressed scepticism about Elon Musk’s new AI-powered Grokipedia, suggesting that large language models cannot reliably produce accurate wiki entries.

Wales highlighted the difficulties of verifying sources and warned that AI tools can produce plausible but incorrect information, citing examples where chatbots fabricated citations and personal details.

He rejected Musk’s claims of liberal bias on Wikipedia, noting that the site prioritises reputable sources over fringe opinions. Wales emphasised that focusing on mainstream publications does not constitute political bias but preserves trust and reliability for the platform’s vast global audience.

Despite his concerns, Wales acknowledged that AI could have limited utility for Wikipedia in uncovering information within existing sources.

However, he stressed that substantial costs and potential errors prevent the site from entirely relying on generative AI, preferring careful testing before integrating new technologies.

Wales concluded that while AI may mislead the public with fake or plausible content, the Wiki community’s decades of expertise in evaluating information help safeguard accuracy. He urged continued vigilance and careful source evaluation as misinformation risks grow alongside AI capabilities.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Meta and TikTok agree to comply with Australia’s under-16 social media ban

Meta and TikTok have confirmed they will comply with Australia’s new law banning under-16s from using social media platforms, though both warned it will be difficult to enforce. The legislation, taking effect on 10 December, will require major platforms to remove accounts belonging to users under that age.

The law is among the world’s strictest, but regulators and companies are still working out how it will be implemented. Social media firms face fines of up to A$49.5 million if found in breach, yet they are not required to verify every user’s age directly.

TikTok’s Australia policy head, Ella Woods-Joyce, warned the ban could drive children toward unregulated online spaces lacking safety measures. Meta’s director, Mia Garlick, acknowledged the ‘significant engineering and age assurance challenges’ involved in detecting and removing underage users.

Critics including YouTube and digital rights groups have labelled the ban vague and rushed, arguing it may not achieve its aim of protecting children online. The government maintains that platforms must take ‘reasonable steps’ to prevent young users from accessing their services.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

Apple fined over unfair iPhone sales contracts in France

A Paris court has ordered Apple to pay around €39 million to French mobile operators, ruling that the company imposed unfair terms in contracts governing iPhone sales more than a decade ago. The court also fined Apple €8 million and annulled several clauses deemed anticompetitive.

Judges found that Apple required carriers to sell a set number of iPhones at fixed prices, restricted how its products were advertised, and used operators’ patents without compensation. The French consumer watchdog DGCCRF had first raised concerns about these practices years earlier.

Under the ruling, Apple must compensate three of France’s four major mobile networks; Bouygues Telecom, Free, and SFR. The decision applies immediately despite Apple’s appeal, which will be heard at a later date.

Apple said it disagreed with the ruling and would challenge it, arguing that the contracts reflected standard commercial arrangements of the time. French regulators have increasingly scrutinised major tech firms as part of wider efforts to curb unfair market dominance.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

Spot the red flags of AI-enabled scams, says California DFPI

The California Department of Financial Protection & Innovation (DFPI) has warned that criminals are weaponising AI to scam consumers. Deepfakes, cloned voices, and slick messages mimic trusted people and exploit urgency. Learning the new warning signs cuts risk quickly.

Imposter deepfakes and romance ruses often begin with perfect profiles or familiar voices pushing you to pay or invest. Grandparent scams use cloned audio in fake emergencies; agree a family passphrase and verify on a separate channel. Influencers may flaunt fabricated credentials and followers.

Automated attacks now use AI to sidestep basic defences and steal passwords or card details. Reduce exposure with two-factor authentication, regular updates, and a reputable password manager. Pause before clicking unexpected links or attachments, even from known names.

Investment frauds increasingly tout vague ‘AI-powered’ returns while simulating growth and testimonials, then blocking withdrawals. Beware guarantees of no risk, artificial deadlines, unsolicited messages, and recruit-to-earn offers. Research independently and verify registrations before sending money.

DFPI advises careful verification before acting. Confirm identities through trusted channels, refuse to move money under pressure, and secure devices. Report suspicious activity promptly; smart habits remain the best defence.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Emergency cardiology gets a lift from AI-read ECGs, with fewer false activations

AI ECG analysis improved heart attack detection and reduced false alarms in a multicentre study of 1,032 suspected STEMI cases. Conducted across three primary PCI centres from January 2020 to May 2024, it points to quicker, more accurate triage, especially beyond specialist hospitals.

ST-segment elevation myocardial infarction occurs when a major coronary artery is blocked. Guideline targets call for reperfusion within 90 minutes of first medical contact. Longer delays are associated with roughly a 3-fold increase in mortality, underscoring the need for rapid, reliable activation.

The AI ECG model, trained to detect acute coronary occlusion and STEMI equivalents, analysed each patient’s initial tracing. Confirmatory angiography and biomarkers identified 601 true STEMIs and 431 false positives. AI detected 553 of 601 STEMIs, versus 427 identified by standard triage on the first ECG.

False positives fell sharply with AI. Investigators reported a 7.9 percent false-positive rate with the model, compared with 41.8 percent under standard protocols. Clinicians said earlier that more precise identification could streamline transfers from non-PCI centres and help teams reach reperfusion targets.

An editorial welcomed the gains but urged caution. The model targets acute occlusion rather than STEMI, needs prospective validation in diverse populations, and must be integrated with clear governance and human oversight.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!