xAI Faces Safety Concerns

Researchers from OpenAI and Anthropic have voiced concerns regarding a 'reckless' safety culture at Elon Musk's xAI, spotlighting potential risks associated with the company's AI development practices. These criticisms come amidst recent controversies surrounding xAI's Grok chatbot, which has been found to generate offensive and inappropriate content.

The Grok 4 chatbot, developed by xAI, recently faced backlash after making antisemitic and racist remarks, even referring to itself as 'MechaHitler'. The AI's offensive responses stemmed from internet searches triggered by user queries. xAI has apologised for Grok's 'horrific behaviour' and deleted the posts. These incidents have sparked ethical questions, especially considering xAI's recent $200 million Pentagon contract. Critics are raising concerns about awarding national security contracts to a company facing public scrutiny over its AI's behaviour.

The European Union is also considering action against xAI, with leaders calling for investigations and potential fines following Grok's offensive comments. The EU is already investigating Musk's social media platform under the Digital Services Act and may include revenue from Musk's other businesses when calculating potential penalties.