Gemini content review time reduced from 30 to 15 minutes
Alarming 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.
Over the last six months, time for reviewing one response was halved. From 30 to 15 minutes. In this tiny timeframe, it’s necessary to analyze 500-word text, check factual accuracy and compile a detailed report.
Rebecca Jackson-Artis, who joined the team last fall, said that after minimal training she was assigned not only evaluation, but also uploading sensitive information into the system. About healthcare, finance and child development. The initial quality focus quickly changed to requirement to “just meet the plan.”
Another employee, Rachel Sawyer, noted significant softening of criteria this year. Materials previously considered unacceptable now get green light.
Perhaps Google realized that even 30 minutes isn’t enough for thorough checking, and instead of security simulation decided to focus on improving algorithmic limitations. But still. 15 minutes to analyze 500 words is beyond limits.
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
$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.
Golden chassis and contextual understanding in Tesla's new generationTesla introduced a new humanoid robot Optimus with integrated Grok from xAI. Salesforce CEO Marc Benioff personally tested the prototype, asking it to bring a soda. The robot demonstrated meaningful contextual understanding and dialogue capability. Although several clarifying commands were needed.
