Slack Faces Backlash for Using Customer Data To Train AI Models

Slack, the cloud-based team communication platform currently owned by Salesforce, faces significant major against its problematic privacy policy regarding AI training data.

The issue gathered the limelight when a Twitter user named Gorey Quinn highlighted a part of Slack’s privacy policy.

The policy states that customer data from Slack global models to train its AI/ML products/services.

The Salesforce-owned company offered only an opt-out system, and customer data was used without proper consent from users.

The privacy policy further adds that workspace owners can send an email request to opt out of the program.

slack signage on an old wall

Following the issue, users have begun opting out of the data-sharing option. Further, users demand better clarification on how Slack uses customer data.

According to its privacy principles page, Slack systems, by default, use customer data like messages, content, and files.

This data is being used in addition to other information, such as usage statistics.

Slack adds that the AI/ML models will be used to provide channel recommendations, search results, emoji suggestions, and autocomplete.

While Slack is not the first company to use customer data for AI training, its appropriation of user consent is problematic. Users are baffled by the idea that their workspace communications are being analyzed.

More importantly, Slack has enabled the option by default instead of asking for consent. The lack of explicit consent in the process is further complicated by inconsistencies in Slack’s privacy policy page.

As there is no way to turn off the option in Workspace settings, owners must email [email protected] to opt out of this process.

Even at the time of writing, the Privacy Principles page says that customer data is being used by AI/ML tools. However, the marketing page for Slack AI claims that the models do not use customer data.

It has also come to light that Slack’s privacy policy does not comply with the EU’s GDPR (General Data Protection Regulation). The policy explicitly states that consent “cannot be implied.” Instead, it “must always be given through an opt-in.”

In response to these concerns, Slack has “updated the language in our Privacy Principles to clarify Slack does not use customer data to train LLMs,” a Salesforce representative told TechWorm.

Per the updated policy, Slack AI does not use customer data to train LLM models. It adds that intelligent features like ranking make use of de-identified user behavior data.

Slack also adds that the LLM models/its own AI/ML tools do not retain customer data.

However, the company has not yet addressed why it has used an opt-out method instead of an opt-in method (as per the GDPR policy.)

Subscribe to our newsletter

To be updated with all the latest news


Please enter your comment!
Please enter your name here

Subscribe to our newsletter

To be updated with all the latest news

Read More

Suggested Post