Shulou(Shulou.com)11/24 Report--

Since artificial intelligence research company OpenAI released AI chat robot ChatGPT in November 2022, the technology giant Amazon has been wary of it, even warning employees not to share confidential information with ChatGPT.

Amazon's internal Slack channel has many employees' questions about how to use ChatGPT. Some employees asked Amazon if it had official guidelines on using ChatGPT on work devices. Others want to know if they are allowed to work with AI tools. An employee urged AWS, Amazon's cloud computing division, to clarify its position on the use of "generative AI (AIGC) tools".

Soon, a corporate lawyer from Amazon joined the discussion. The lawyer warned employees not to provide "any Amazon confidential information" to ChatGPT, including Amazon code being written, according to screenshots of internal communication on Slack. He also advises employees to follow the company's existing confidentiality policy because some of ChatGPT's responses look very similar to Amazon's internal situation.

"this is crucial because your input may be used as iterative training data for ChatGPT, and we do not want its output to contain or similar to our confidential information," the lawyer wrote. "

These exchanges show that the sudden emergence of ChatGPT has given rise to many new ethical issues. ChatGPT is a conversational AI tool that can respond to queries with clearer and more intelligent answers. The rapid spread of ChatGPT has the potential to upend many industries, including media, academia and healthcare, prompting efforts to find new use cases for chatbots and their possible impact.

How employees share confidential information with ChatGPT, and how its developer OpenAI handles it, can be a thorny issue. This is particularly important for Amazon because its main competitor Microsoft has invested heavily in OpenAI, including a new round of financing this week, which reportedly totaled $10 billion.

Emily Bender, who teaches computational linguistics at the University of Washington, said: "OpenAI is far from transparent about how to use data, but if the data is used for training, I expect companies to think: after a few months of widespread use of ChatGPT, is it possible to obtain confidential information from private companies through well-designed tips?"

Amazon has put in place a number of internal protection measures for employees to use ChatGPT. Screenshots of information exchanged, for example, show that when employees use work devices to access ChatGPT, a warning message pops up that they are about to access a third-party service that "may not have been approved by Amazon security".

Employees who participated in the Slack chat said they could bypass the message simply by clicking on the "Acknowledge" tab. Staff speculate that the warning pop-up window is to prevent employees from pasting confidential information on ChatGPT, especially because they have not yet seen the company's policy on internal use.

Still, some Amazon employees are already using AI tools as software "coding assistants" to help improve internal lines of code. "I think it would be great to be able to use this feature directly now! so any guidance would be great," an Amazon employee wrote on Slack.

Worried and excited, another employee said he shared Amazon's interview questions for an open programming position on ChatGPT. According to the employee's Slack post, the AI model provides the right solutions to several of these technical problems. "I am both worried and excited to see how this will affect the way we conduct programming interviews," he said. "

Overall, Amazon employees who chat on Slack are excited about the potential of ChatGPT and wonder if Amazon is developing competing products. Corporate lawyers who warn employees not to use ChatGPT say Amazon is accelerating the development of "similar technologies", citing voice assistant Alexa and code recommendation service CodeWhisperer.

One AWS employee wrote that the Enterprise Support team recently set up a small internal working group to "understand the impact of advanced chat AI on our business." Research shows that ChatGPT is "very good" at answering AWS support questions, including solving Aurora database problems. At the same time, it is also excellent at creating training materials for AWS-certified cloud architect exams and proposing customers' corporate goals.

Bender, of the University of Washington, says the increasing use of ChatGPT at work raises serious questions about how OpenAI plans to use materials shared with AI tools.

OpenAI's terms of service require users to agree that they can use all inputs and outputs generated by users and ChatGPT. It also indicates that OpenAI removes all personally identifiable information (PII) from the data used.

But given the rapid growth of ChatGPT, which topped 1 million users within a week of launch, it's hard to see how OpenAI can identify and delete all personal information, Bender said. More importantly, the company's intellectual property may not be part of the PII definition.

For Amazon employees, data privacy seems to be the least concern. They say the use of chatbots at work has increased productivity tenfold, and many people want to join internal teams that develop similar services.

One of the employees wrote on Slack: "if there are plans to build similar services, I would very much like to join them and make a modest contribution when necessary."

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.