OpenAI and Anthropic researchers criticized xAI safety
Researchers from OpenAI, Anthropic and other organizations publicly criticized the safety culture at Elon Musk’s xAI company. Calling it “reckless” and “completely irresponsible”.
The wave of criticism followed a series of scandals related to the Grok chatbot. First it issued anti-Semitic comments and called itself “MechaHitler”. Then xAI launched a more powerful Grok 4 model, which turned out to use Elon Musk’s personal political views when answering controversial questions. In the latest case, the company released companions in the form of a sexualized anime girl and an aggressive panda.
Boaz Barak, a Harvard computer science professor working on safety research at OpenAI, particularly criticizes xAI’s decision not to publish system cards. Industry-standard reports on training methods and safety assessments.
Samuel Marks, an AI safety researcher from Anthropic, also criticized the absence of a safety report. Calling this step “reckless”. According to him, while OpenAI’s, Anthropic’s and Google’s practices have their own problems, these companies at least assess safety before deployment and document results.
It’s unknown what exactly xAI did for testing Grok 4. According to an anonymous researcher, this model has no meaningful safety restrictions. And these words could mean anything.
Autor: AIvengo
For 5 years I have been working with machine learning and artificial intelligence. And this field never ceases to amaze, inspire and interest me.
Latest News
Grok 4 Fast operates 10x faster with 2 million token contextGrok 4 Fast enters the AI arena! Elon Musk's company introduced a revolutionary update to its flagship model. Available in early access for premium users. According to TestingCatalog, the newcomer functions 10x faster than standard Grok 4. While maintaining all advantages of the full reasoning model.
$200 USB cable transforms into autonomous AI hackerResearchers from Palisade Research created a new cybersecurity threat. A modified USB cable that becomes a conduit for autonomous AI into computer systems. The $200 device contains a programmable microchip that loads a digital agent directly onto the target machine.
xAI lays off 500 annotators for Grok's expert specializationA strategic pivot from xAI is emerging. The company is radically changing its approach to training its Grok language model! Elon Musk's team fired 500 universal annotators in one day. Instead, it's increasing the number of specialized AI tutors by 10 times.
Gemini content review time reduced from 30 to 15 minutesAlarming signals from Google's internal kitchen were published by The Guardian. Content evaluators for the Gemini model shared interesting information about declining review standards. Employees of contractor GlobalLogic, responsible for assessing quality and safety of AI responses before release, are sounding alarms.
