Increasing numbers of workers in the United States are embracing ChatGPT, a conversational AI program, to assist with routine tasks, according to a recent Reuters/Ipsos poll. Despite apprehensions that have prompted companies such as Microsoft and Google to impose restrictions, the adoption of ChatGPT is gaining momentum across industries.
The spotlight on ChatGPT arises from its utilization of generative AI to engage in conversations and address diverse prompts. While organizations worldwide explore ways to harness its potential, security firms and corporations are voicing concerns about the potential leakage of intellectual property and strategic information.
Numerous anecdotal accounts highlight the practical applications of ChatGPT in day-to-day work, including drafting emails, summarizing documents, and conducting preliminary research. Remarkably, 28% of the survey respondents in an online poll focused on artificial intelligence (AI), conducted between July 11 and 17, admitted to regularly incorporating ChatGPT in their work routines. Strikingly, only 22% of participants affirmed that their employers explicitly endorsed the use of external tools like ChatGPT.
The survey, encompassing 2,625 adults across the U.S., carried a credibility interval—a measure of precision—of approximately 2 percentage points. Out of those polled, 10% confirmed that their employers prohibited the use of external AI tools, while around 25% remained uncertain about their company’s stance on the technology.
The rapid ascension of ChatGPT since its November launch has engendered both excitement and apprehension. Developer OpenAI has encountered regulatory challenges, particularly in Europe, where privacy watchdogs have raised concerns about data collection practices. Moreover, concerns have emerged about the potential for AI to inadvertently reproduce sensitive data absorbed during training, raising proprietary information risks.
The lack of clarity surrounding the use of generative AI services and the potential for data misuse have raised alarm among experts. Ben King, VP of Customer Trust at corporate security firm Okta (OKTA.O), stated, “People do not understand how the data is used when they use generative AI services.” He added that businesses face challenges as users often lack contractual agreements with these free AI services, making conventional risk assessment processes inadequate.
OpenAI declined to comment on individual employees’ use of ChatGPT but emphasized in a recent blog post that corporate partners’ data would not be used to further train the chatbot without explicit permission.
The growing adoption of ChatGPT is not without its complexities. Concerns about data security, intellectual property, and proprietary information underscore the need for companies to tread carefully as they integrate AI solutions into their workflows. The diversity in organizations’ approaches—ranging from outright bans to cautious experimentation—reflects the ongoing efforts to harness AI’s benefits while safeguarding sensitive information.
As the AI landscape evolves, businesses are grappling with striking the right balance between productivity enhancement and information security. The measured approach is underscored by the potential risks, as experts advise careful consideration to ensure that the benefits of AI are harnessed responsibly.