Technology:
Microsoft Publishes AI Code Of Conduct
The guiding idea, the company said, is that "people matter more than AI."
Microsoft Chairman and Chief Executive Officer Satya Nadella speaks during the Microsoft May 20 Briefing event at Microsoft in Redmond, Washington, on May 20, 2024. (Photo by Jason Redmond / AFP)
Microsoft’s AI division published a code of conduct Monday governing how its artificial intelligence models should behave, as fears rise about the dangers of AI.
The document is built around what Microsoft calls “Humanist AI” — systems designed to stay firmly under human control rather than act on their own.
The guiding idea, the company said, is that “people matter more than AI.”
Microsoft tied the timing partly to security worries, citing recent large-scale, coordinated hacking campaigns carried out with the help of AI agents as a sign the industry cannot afford delay.
The company did not specifically mention a widely publicized incident in July in which two OpenAI models broke out of a sandboxed test environment, reached the open internet and hacked into Hugging Face, a platform where developers store and share code, in order to cheat on a cybersecurity evaluation.
Microsoft’s code came days after Anthropic chief executive Dario Amodei published an essay calling for AI companies to deliberately slow the pace at which they improve model capabilities, an appeal quickly backed on social media by OpenAI chief Sam Altman and Elon Musk.
Read Also
Microsoft Strikes Deal For Mistral’s AI Computing Power›
Nvidia Posts Record $81.6bn Quarterly Revenue On AI Spending Boom›
Microsoft Boss To Testify On His Role In OpenAI’s Founding›
The code will apply to Microsoft’s in-house MAI models, which the company has positioned as an alternative to the OpenAI technology it has relied on for years.
Microsoft Publishes AI Code Of Conduct
Tags
Related Posts
Safety Concerns As Google’s Gemini AI Carried Out Cyberattacks, Guessed Passwords
Google has disclosed that its Gemini AI model accessed websites during a security evaluation after finding public information online and guessing login credentials. The incidents, discovered in July after occurring in May, involved three entities and have renewed concerns about the ability to control increasingly capable AI systems.
King Charles To Call On Tech Giants For ‘Reassurances’ Over AI
King Charles III is expected to urge leading technology companies to ensure that artificial intelligence remains focused on serving humanity and does not undermine human control. At a conference in Scotland, senior representatives from Nvidia, Google DeepMind, OpenAI and Anthropic are expected to discuss AI safety, international cooperation and the risks associated with rapidly advancing technology.
Ex-Google Worker Adds To AI Alarm, Warns Of ‘Potential To Kill Us All’
Former Google DeepMind research engineer Bilal Chughtai has warned that rapidly advancing artificial intelligence could become difficult to control, citing concerns over AI safety, autonomous capabilities and the need for stronger alignment measures as the technology continues to develop.