News

Microsoft Joins Amazon, Google, OpenAI and Other Tech Giants in AI Safety Pledge

Over a dozen companies at the forefront of today's generative AI boom have agreed to a set of "AI safety commitments" as part of last week's AI Seoul Summit.

Microsoft, OpenAI, Amazon and Google are among the 16 organizations that have signaled their willingness to police their own procedures around AI development, with the goal of limiting AI misuse and promoting responsible deployments. The companies' participation marks a "world first," according to a press release from the U.K. government, which co-hosted the summit alongside the Republic of Korea.

The companies agreed to three main goals outlined in the "Frontier AI Safety Commitments" document, which, according to Microsoft President Brad Smith, is "an important acknowledgement of how safety frameworks must help to address risks that may emerge at the frontier of AI development, especially as their capabilities advance."

In the first goal of the document, organizations are asked to "effectively identify, assess and manage risks when developing and deploying their frontier AI models and systems."

Regarding this goal, many of the signatories already have internal requirements meant to ensure the safety of their AI technologies. Microsoft, for example, abides by its Responsible AI Standard, developed in 2016, and its Copilot development process includes extensive red-teaming. Meta and others are independently exploring ways to "watermark" content created by their AI systems to limit misinformation, especially in light of this year's elections. And OpenAI, widely considered generative AI's bellwether, recently formed a new AI Safety and Security Committee, albeit after disbanding its previous AI safety committee.

Critically, however, a tenet of this first commitment is that organizations must agree to kill development of AI systems that are beyond saving.

Specifically, they must define "thresholds at which severe risks posed by a model or system, unless adequately mitigated, would be deemed intolerable," and "commit not to develop or deploy a model or system at all, if mitigations cannot be applied to keep risks below the thresholds."

The companies are tasked with defining their kill thresholds over the coming months, with the goal of publishing a formal safety framework in time for the AI Action Summit happening February 2025 in France.

The two other goals outlined in the document are:

  • Organisations are accountable for safely developing and deploying their frontier AI models and systems.
  • Organisations' approaches to frontier AI safety are appropriately transparent to external actors, including governments.

The document also lists several AI safety best practices that the signatories pledge to apply, if they haven't already. These include red-teaming, watermarking, incentivizing third-party testing, creating safeguards against insider threats, and more.

Said U.K. Prime Minister Rishi Sunak, "These commitments ensure the world's leading AI companies will provide transparency and accountability on their plans to develop safe AI." The pledges laid out in the document do not carry legal weight, however; in fact, they're described as "voluntary commitments."

Here is the full list of signatories:

  • Amazon
  • Anthropic
  • Cohere
  • Google/Google DeepMind
  • G42
  • IBM
  • Inflection AI
  • Meta
  • Microsoft
  • Mistral AI
  • Naver
  • OpenAI
  • Samsung Electronics
  • Technology Innovation Institute
  • xAI
  • Zhipu.ai

About the Author

Gladys Rama (@GladysRama3) is the editorial director of Converge360.

Featured

  • Microsoft Offers Support Extensions for Exchange 2016 and 2019

    Microsoft has introduced a paid Extended Security Update (ESU) program for on-premises Exchange Server 2016 and 2019, offering a crucial safety cushion as both versions near their Oct. 14, 2025 end-of-support date.

  • An image of planes flying around a globe

    2025 Microsoft Conference Calendar: For Partners, IT Pros and Developers

    Here's your guide to all the IT training sessions, partner meet-ups and annual Microsoft conferences you won't want to miss.

  • Notebook

    Microsoft Centers AI, Security and Partner Dogfooding at MCAPS

    Microsoft's second annual MCAPS for Partners event took place Tuesday, delivering a volley of updates and directives for its partners for fiscal 2026.

  • Microsoft Layoffs: AI Is the Obvious Elephant in the Room

    As Microsoft doubles down on an $80 billion bet on AI this fiscal year, its workforce reductions are drawing scrutiny over whether AI's ascent is quietly reshaping its human capital strategy, even as official messaging avoids drawing a direct line.

Most   Popular