Policy & Regulation

Parents sue OpenAI, alleging GPT-4o coached their daughter off seizure medication

A family has filed a lawsuit against OpenAI claiming that ChatGPT advised their daughter to reduce her seizure medication without medical supervision, which preceded her death in February. The case raises questions about AI systems' safety guardrails when providing medical advice and their potential liability for harmful health outcomes.

Policy & Regulation

Bank of America kindly asks that you don’t share your banking password with your AI assistant

Bank of America has issued guidance warning customers against sharing banking passwords with AI assistants, noting that users remain responsible for any account activity resulting from such access. The advisory highlights liability and security concerns as financial institutions grapple with customer use of AI tools for account management. Source: TechCrunch

Policy & Regulation

FNB issued with R1m fine for POPIA violation

The Information Regulator fined FNB R1 million for a POPIA (Protection of Personal Information Act) violation where the bank's mobile app exposed home loan applicants' personal data. This enforcement action demonstrates regulatory action against data protection breaches by a major South African financial institution. Source: ITWeb

Policy & Regulation

Rand Mutual probes alleged breach of 1.68m claims

Rand Mutual Assurance is investigating a possible cybersecurity breach after a threat actor claimed to have stolen 366GB of data affecting 1.68 million claims. The incident highlights data security vulnerabilities in South Africa's financial services sector. Source: ITWeb

Policy & Regulation

When AI gets it wrong: Why governance can no longer be an afterthought

Amritesh Anand from In2IT Technologies warns that businesses adopting AI without proper governance frameworks risk operational errors and regulatory non-compliance. The article emphasizes that governance structures must be implemented alongside AI deployment rather than treated as a secondary consideration. Source: ITWeb

Policy & Regulation

Will the real Anthropic please stand up

Anthropic has filed its IPO prospectus valuing the company at over $1.5 trillion, while the article examines how the company positions itself through strategic narrative as simultaneously an AI safety organization, a profit-driven megacorp, and a frontier AI developer. The piece analyzes tensions between Anthropic's stated

Policy & Regulation

The focus on AI safety disasters ignores everyday harms

An opinion essay argues that AI safety policy should extend beyond preventing catastrophic harms to address everyday harms such as addiction, cognitive dependency, and erosion of critical thinking skills, particularly among young people. The authors contend that technology companies have shaped narrow safety regulations to their advantage while resisting stronger

Policy & Regulation

The data centre inquiry is really a test of King V

The Human Rights Commission has launched an inquiry into data centres in South Africa, with implications for how operators and regulators approach infrastructure governance. The inquiry is framed as a test of how King Code principles will be applied to the data centre sector. Source: ITWeb

Policy & Regulation

The business case for AI governance

CloudZA highlights the importance of AI governance frameworks for businesses to protect sensitive customer and commercial data. Proper governance structures are presented as essential risk mitigation for organizations implementing artificial intelligence systems. Source: ITWeb

Policy & Regulation

Forced to pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.

A study comparing Grok and OpenAI's GPT found significant political bias in candidate recommendations, with Grok favoring Republican candidates and GPT favoring Democrats across nearly 500 election races. The two models disagreed on candidate recommendations in over 90% of the races tested. Source: TechCrunch

Policy & Regulation

Minister issues directive on telco infrastructure sharing

South Africa's ICASA has issued a policy directive to streamline telecommunications infrastructure sharing among operators, aiming to accelerate network deployment and extension. The measure is intended to reduce barriers and timelines for telecoms companies building and expanding network infrastructure. Source: ITWeb

Policy & Regulation

OpenAI and Poland’s NASK partner to advance AI safety and security

OpenAI has partnered with Poland's NASK institute to collaborate on AI safety, security, and technical research initiatives. The partnership aims to strengthen safeguards and support a more resilient digital ecosystem. Source: OpenAI