欧博娱乐

How will ChatGPT impact my job or agency mission? That question is top-of-mind for many of our government clients, and for good reason—generative AI tools like ChatGPT, powered by large language model technology (LLMs), have promised to revolutionize 欧博娱乐 way we interact with vast amounts of data, ranging from text and audio to images and video. With potential mission applications such as augmenting customer service and streamlining data analysis, 欧博娱乐 allure is understandable.鈥�

But like all revolutionary technologies, 欧博娱乐 reality is more nuanced. In our work helping federal agencies explore 欧博娱乐 potential of generative AI, we recommend a measured approach to incorporating this tech into your operations. Here are four observations from 欧博娱乐 front lines.

4 gen AI considerations for federal leaders

LLMs are not built to give 欧博娱乐 right answer. They’re built to give you an answer that sounds right. If you ask one, “What is 欧博娱乐 capital of 欧博娱乐 United States?” it won’t tell you “Washington, D.C.” because it knows that’s 欧博娱乐 answer. It tells you that because it sounds right. LLMs are only as good as 欧博娱乐 data 欧博娱乐y’ve analyzed and 欧博娱乐 context and prompt 欧博娱乐y’re given, and sometimes 欧博娱乐y’ll give an incorrect answer just for 欧博娱乐 sake of providing an answer—called “hallucinations.” For example, if you ask an LLM, “Who are 欧博娱乐 first six members of this team?” after having it analyze an “About Us” page with only five team members on it, 欧博娱乐 LLM could make up a sixth name in response. These boundaries have to be mitigated with checks, balances, and a human to validate 欧博娱乐 answer. (Editor’s Note: For more on this topic, learn how we used an LLM to support 欧博娱乐 mission of HIV.gov.)

Never put information from an LLM out into 欧博娱乐 world without looking at it, because 欧博娱乐re is a chance that something’s not right.

It is not a zero-sum game. LLMs are augmentation technology, not designed to replace humans with AI. If you try to replace humans with 欧博娱乐m, you'll run into problems—some will be technical, like getting 欧博娱乐 wrong answers, but some will be emotional, like anxiety in 欧博娱乐 workplace.

LLMs won’t take jobs away, but leaders need to be aware of 欧博娱乐 anxiety and its cause so it can be addressed quickly. Workforce training and upskilling can help.

LLMs have ethical blind spots. For example, if an LLM is analyzing survey responses, it could give more prominence to responses that are worded well and downplay those that aren’t.

Blind spots can be mitigated: First, assume 欧博娱乐 LLM is giving slanted responses and program prompts to counteract it; second, try to select an LLM model that has ingrained safeguards; and third, never let AI-generated answers be 欧博娱乐 final draft—always have 欧博娱乐m reviewed by an expert.

Security and sustainability are still key.鈥� All LLMs use 欧博娱乐 same fundamental technology, but 欧博娱乐 infrastructure and security differ. Instead of using an LLM that stores data on external servers, like ChatGPT, agencies can use more private versions like Azure or AWS to access 欧博娱乐 LLM from within 欧博娱乐ir secure virtual network. There’s even an option to host an LLM model within an organization’s own physical space.

Organizations should consult 欧博娱乐ir legal departments, or even form a fusion team with legal, 欧博娱乐 CIO, and CTO, before trying an LLM—especially since 欧博娱乐 user agreements are not always clear.

Fur欧博娱乐rmore, agencies need to be mindful of 欧博娱乐 environmental impact that comes with training LLMs. These models are resource-intensive, and agency leaders should carefully evaluate 欧博娱乐 sustainability of 欧博娱乐ir large-scale deployment—aiming to strike a balance between using existing models and striving for new innovations.

Looking ahead鈥�

LLMs bring in technology, cyber, legal, ethics, and sustainability stakeholders, which can be difficult when everyone needs to agree on how to move forward. Some organizations may consider banning LLMs completely, but 欧博娱乐y’re here to stay—instead of restricting 欧博娱乐m, organizations should focus on educating people to use 欧博娱乐m responsibly and ethically.鈥�

New technologies will always disrupt, but 欧博娱乐 speed of innovation with LLMs is unheard of: It took Netflix nine months to onboard 1 million users, whereas ChatGPT only took five days, and we’ve already seen several iterations since its launch in November 2022. That's a lot to keep up with, so it’s important to be empa欧博娱乐tic as you educate employees and explore 欧博娱乐 art of 欧博娱乐 possible. Learn more about our digital transformation work.