Slack responds to online criticism by clarifying its data policy regarding AI usage

  • Slack clarifies data policy amid online criticism regarding AI usage.
  • Platform assures users that AI models do not memorize or reproduce customer data.
  • Company acknowledges use of machine learning for specific features.
  • Blog post emphasizes commitment to transparency and revising privacy principles.
  • Users provided with insights into types of customer data utilized.
  • Option available for users to opt out of data usage.
  • Concerns persist regarding AI and data privacy in the market.

Main AI News:

In response to recent online criticism, Slack, the renowned workplace communication platform, has taken steps to clarify its data policy, particularly concerning artificial intelligence (AI) usage. Amid allegations of ambiguity surrounding its data practices, Slack, owned by the US tech giant Salesforce, faced scrutiny from social media users. The platform witnessed a surge in adoption during the COVID-19 pandemic, as remote work became prevalent.

The controversy arose when a user expressed surprise at Slack’s privacy principles, revealing that customer data, including messages, content, and files submitted to the platform, are subject to analysis for developing non-generative AI and machine learning. Addressing these concerns, Slack emphasized that its AI models are not designed to memorize or reproduce customer data, nor are they utilized to train third-party large language models (LLMs). However, the company acknowledged employing machine learning for specific features like summarization.

Responding to the outcry, Slack released a blog post earlier this month, acknowledging community feedback and pledging greater clarity in its privacy principles. While reaffirming its commitment to data privacy, Slack underscored the importance of distinguishing between traditional ML models and generative AI in its data utilization approach.

In an exclusive interview with Euronews Next, Slack provided insights into the types of customer data used, including message timestamps, user interactions, and channel relevance indicators. Although the guidelines remained unchanged, the company revised the wording of its confidentiality principles to offer better transparency regarding its practices and policies.

Furthermore, Slack clarified that its traditional ML models operate on de-identified, aggregate data and refrain from accessing message content in direct messages (DMs), private channels, or public channels. Notably, users have the option to opt out of data usage by contacting customer service via email, demonstrating Slack’s commitment to user autonomy.

Nevertheless, concerns persist regarding the intersection of artificial intelligence and privacy. While AI holds the potential to streamline operations, apprehensions surrounding data privacy, particularly in the context of large language models, continue to mount. Instances of tech giants cautioning employees against sharing confidential information with generative AI tools further underscore the evolving landscape of data privacy in the digital era.


Slack’s proactive approach to addressing concerns about its data policy reflects a commitment to transparency and user privacy. By clarifying its practices and offering opt-out options, Slack aims to instill confidence among users and maintain its position as a trusted workplace communication platform. This emphasis on clarity and privacy could influence market dynamics, as consumers increasingly prioritize transparency and data protection in their technology choices.