Samsung has banned its staff from using ChatGPT and other generative AI tools on internal networks and company-issued devices. This decision comes after three employees shared sensitive data with ChatGPT in April. Samsung warned its staff to cease using AI tools, or they could face disciplinary action, including dismissal.
According to Bloomberg, a memo circulating at Samsung’s headquarters in Suwon on Monday notified staff about the ban. A Samsung representative told the news outlet that the company communicated the decision with its staff last week.
While generative AI tools like OpenAI’s ChatGPT and Google’s Bard can boost productivity in the workplace, they also raise privacy concerns. The information shared in prompts, which may include confidential data, is used to train these algorithms, and there’s the ever-present possibility that this data could fall into the wrong hands.
In February, after analyzing how 1.6 million workers at different companies use ChatGPT, Cyberhaven Labs revealed that 11 percent of the data employees share with the chatbot is sensitive information.
Samsung Staff Shared Confidential Data With ChatGPT
Last month, local media reported that three engineers at the South Korean company shared source code with ChatGPT. They allegedly asked ChatGPT to suggest improvements to the confidential code. The three employees are being investigated, Samsung said.
Samsung staff also shared general internal information, information about equipment, and meeting content with the chatbot, presumably as staff attempted to improve code and use ChatGPT to summarize data.
Upon realizing this, Samsung restricted the length of ChatGPT prompts for its staff to just 1024 characters, local media reported.
“HQ is reviewing security measures to create a secure environment for safely using generative AI to enhance employees’ productivity and efficiency,” Samsung’s memo said, adding that it is “temporarily restricting” the use of AI tools.
Samsung is also considering launching a proprietary AI service via its Innovation Center.
ChatGPT and Growing Privacy Concerns
Samsung’s decision to ban generative AI tools reflects a growing concern in the corporate world. Since the launch of ChatGPT, several top companies, including JP Morgan, Amazon, Bank of America, and Verizon, have imposed restrictions on the use of generative AI tools in their workplaces over privacy concerns.
To address these concerns, OpenAI recently added an “incognito mode” to its popular chatbot. When this feature is activated, the company doesn’t store users’ conversations to train its chatbot. OpenAI is also working on additional data controls for businesses.
You can disable history and model training on ChatGPT by going to Settings > Data Controls, and toggle Chat History & Training off. If you are in the EU, you can also get OpenAI to delete your data by filling out this form.
If ChatGPT is blocked in your country, you can learn how to unblock the chatbot from anywhere in our guide to accessing ChatGPT.
