With OpenAI ChatGPT™, large language models (LLMs) got their iPhone moment. Now, everyone can experience the power of LLMs and understand their usefulness in our everyday lives. The hype around LLMs is just the beginning. This technology is already transforming many industry services.
For service management, generative AI and LLMs are poised to reduce the burden on tier 1 support agents. Finally, virtual agents—now powered by LLM—can live up to and even exceed user expectations. And organizations can create more efficient, effective, and personalized employee interactions, ultimately improving the user experience and tool adoption.
Private, generative AI virtual agent in SMAX
We are excited to announce the technical preview of a private, generative AI virtual agent for OpenText™ IT Service Management (ITSM). The new virtual agent, ITOM Aviator is powered by a secure LLM and connects across your enterprise content to deliver human-like, contextually relevant responses to user requests. Without relying on support or compromising enterprise data privacy, users can get the answers they need, when they need them. This way everyone can get back to work faster, and productivity soars as a result.
Aviator demonstrates connected enterprise intelligence at work. For example, HR managers can ask Aviator about company-designated holidays for their home countries. Procurement administrators can request contract details such as contract value, approval date, and associated risks. And employees can discuss available models for a new laptop and quickly place their order.
Stay secure with a private LLM
Public generative AI raises data security and privacy concerns for enterprises. After all, public generative AI gives the public access to your questions and answers. And are public generative AI answers correct (even if they may seem correct)? Trust in LLM is important. That’s why we’ve made our LLM private.
Aviator, the OpenText private generative AI virtual agent, delivers:
- Privacy and security You have complete control over the model’s architecture, training data and training process. OpenText hosts and operates the LLM privately, ensuring integrity of your data and answers from the virtual agent.
- Enterprise intelligence. The virtual agent accesses enterprise data and documents across your content management system and answer questions in real-time for any enterprise function. It can also handle large documents.
- Access control. The virtual agent provides answers and information based on employee role, location, and group membership. You can tag information and documents with entitlement labels. The user context and data entitlements enforce real-time access control filtering to make sure answers are relevant and permissible.
- Ubiquitous Interface. The virtual agent is available from a variety of interfaces, including the SMAX service portal, Microsoft Teams, and via a public API.
OpenAI ChatGPT-connector for service desk agents
With the CE 23.3 SMAX release, the OpenAI API connector is generally available to help service desk agents troubleshoot technical issues and provide well-crafted conversational answers based on information that’s publicly accessible. You can read more about this integration on the SMAX Practitioner Portal.
Just as AI has permeated our personal lives and ChatGPT has captured our imaginations, generative AI has a role to place in the workplace too. You can learn more about what it takes to build a game-changing, generative AI virtual agent in this blog. The OpenText private generative AI virtual agent can help move your service management into the future with these benefits:
- Secure with private LLM: Keep confidential and intellectual property securely inside your own network.
- Lower costs: Reduce the burden on tier 1 service desk agents and automate smart routing of tier 2 escalations.
- Better employee experiences: Engage employees with natural language and smart self-service.
- Faster issue resolution: Smart contextual search provides incident resolution steps and automation fulfils requests.