How are you starting to use AI for work?

I think the AI revolution is absolutely amazing.  Professionally I've used ChatGPT to overcome writer's block on blog posts, troubleshoot niche answers to highly technical questions and generate new ways of thinking about old problems. 

On the Zendesk side, I've seen what the power of Zendesk Advanced AI can do, from generating help center content to summarizing tickets, we are just scratching the surface on how AI can help agents gain efficiency.

Outside of work, I've 100% leveraged ChatGPT to help craft custom bedtime stories for my kids.

@brandon12 Not going to lie, I love the idea of using ChatGPT for custom bedtime stories and would have never thought to do that! I'm going to start doing that :) 

The generative AI has been useful along with summarizing.  Our team has historically used shortcuts over macros which makes suggested macros useless.  

What I am curious to learn, is how the Generative AI is going to do against the Quality Assurance AI features that are starting to appear as offerings. What if they do not get along? 

GenAI will help our agents to decrease workload in the daily business. We are planning to run the extended AI this year, first we had to challenge the use case with IT security. :-)

In fact, the use case of using prompts for messages, receiving a summary, intents etc. and to use generative AI to issue articles: It is a game changer!

@steve25: All the challenges connected to AI have to be considered as well. Absolutely correct.

Generative AI is something we are interested in, but we aren't ready to make the leap yet. Implementing AI as a B2B company has some additional challenges, though things such as conversation summaries and suggested Guide articles for common questions are a couple of ways that I see generative AI being very useful for our organization. 

Before we dive into AI, we want to make sure we implement the many awesome tools we already have. From a CX perspective, AI can really enhance the experience, but only if it is implemented correctly. With the pressure for efficiency, it's important to balance efficiency with the quality of the customer interaction. Using AI to gauge customer intent or to proofread responses (tone-shift) can have a real impact on helping agents increase the quality of their work and enhance their connection with the customer. 

I personally have used generative AI tools such as ChatGPT very little, but have had colleagues use it in a similar manner to what @brandon12 mentions. Using ChatGPT to get over writer's block is a great use of the tool and using it to help generate content from a group of ideas can really save a lot of time. 

Something big we did last year was enable a ChatGPT component in our chatbot. We are a lean team engaging with 10k+ users every month. And being focused on education, our chatbot suffers from a high fallback rate due to what appears to be children asking or saying odd/irrelevant things. Enabling ChatGPT in the bot has dramatically lowered our fallback rate and gave a huge boost to our self-service ratio.

Important perhaps to note that we have not linked our chatbot to any user data for privacy reasons. Conversations are purely anonymous. So having double-digit KPI improvements with only public help center content is crazy.

While it's great that we're seeing fewer support tickets, we're now seeing a higher percent of complex technical issues. So we're having to somewhat reinvent how we assess our manpower needs, as we feel a need for more manpower to help stay on top of fewer tickets. We need to find a way to show with data that tickets are overall more complex. This requires greater granularity than what we've been tracking, which leads to a new area where AI may soon prove useful to us.

@bryan18 That sounds amazing! 👆 Interesting problem of having fewer tickets of higher complexity.

One of the biggest wins last year was having GPT act as my tutor when I got into Zendesk Integration Services (ZIS). I'm not a developer and it was awesome to have suggestions on how to design a flow, overcome obstacles, evaluate alternative approaches, debug failures, etc. - no human would have put up with me in these sessions, but GPT came through most of the time and managed to create several flows we put in production.

I also used it for summarizing long texts, meeting transcripts, etc., and outputting data in a certain format, like the JSON body for updating schedules using the API.

I expect to find even more uses this year and also build our own ticket summary app.

Remember to say please!

That reminds me of another thing! Zendesk now has a beta going on to test some new ticket deletion tools. But before this was going on, I used GPT to help me create a google apps script that bulk deletes tickets. Since zendesk API docs are public, GPT was able to compose a script that required minimal intervention to get it working.

In combination with the google sheets zendesk connector extension, I created a spreadsheet that fetches up to 1000 tickets each day that are 2 years old and lists only the ticket IDs. Then, the google apps script runs every 2 hours deleting up to 100 tickets. This helps us stay in compliance with GDPR automatically.

Awesome use-cases everyone. Keep them coming! Since it's relevant to this conversation I wanted to share that @eric28 just hosted the first AI User Group Meetup which you can find a summary of here: A.I. User Group - January 8, 2024 - Recap and Continued Discussion

This session focused on using AI for:

• Enhancing Knowledge Bases
• Low Touch Tickets
• Understanding Customer Sentiments

I've been playing with putting our communications (with redacted info) through a few different AI chats and the results have been great (ChatGPT -> Bard -> Zendesk for kicks). Just like anything else, you need to take a critical look at the results, but I've been able to gather new ways of communicating difficult ideas or opinions that we are slowly starting to put into our every day use.

Hi everyone,

Thanks for sharing how your teams are leveraging AI! It's really great to see the creative ways companies are implementing this technology.

At our company, we've been continuing to enhance our AI journey and are excited about new possibilities. So far, we have started using AI for:

Analytics: We are feeding historical chat transcripts into an AI analytics platform to uncover insights about common customer questions, pain points, and suggestions. This is helping guide our self-service content strategy.

Ticket Summarization: For especially long or complex tickets, we are testing an AI summary tools from Tethr that digests the key details into a concise summary. This helps agents process the details more efficiently.

Intent Detection: We have started to use Zendesk AI to analyze visitor questions and predict their intent. Based on the intent, we've created many autoreply bots that can provide tailored responses to common inquiries.

Future Initiatives: Our team is exploring ways to enhance our chatbot capabilities further using generative AI. We are also utilizing and enhancing our AI-powered quality assurance review tool to analyze agent tones and messages.

As we continue our AI adoption, we hope to learn from the community's experiences here and love to see everyone's comments.

Planning to use AI to build a chatbot.

This is awesome. Thanks for sharing @dakota12 and @lord,. It's exciting to see how everyone is using AI to improve the customer and employee experience :) 

We are also experimenting with the many of the mentioned AI functionalities with security and feasibility in mind. 

In addition, just to try it, we've used Synthesia to make an AI avatar presentation to summarize our information security awareness training messages as a wrap up to our awareness campaign. It was a great success, so I'm contemplating on how we can introduce this technology for employee training in general as an alternative to emails, meetings and slides. 

Hi, this is Lifan from Aissist.io. We utilize generative AI like GPT to automate sales and service queries. We are able to help our customers fully automate 80% of their queries. Not just draft or summarize, but full automation. Please contact us if you wanna a demo.

We're using AI to help reply to users and generate draft replies for our agents. We're an AI customer support Zendesk marketplace app (that we built ourselves) for our agents.

It's saving everyone a few hours a week and currently deflecting 70%+ of our support tickets. Which is great as it frees up agents for the more complex tickets.

I'm seeing more and more companies adopt AI in customer support, which is super exciting!

I always love this group: https://support.zendesk.com/
They are so updated and has best product in the market. The products which did not update themselves, they are now obseleted. I am also looking forward to enhance my experience in AI by using zendesk.com produtcs and support.

We are using AI insights to monitor our calls with the integration with CloudTalk.

Features such as Sentiment Analysis, Trending Topics, Call Scoring, and Talk/Listen Ratio provide managers with deep insight into customer behavior and agent performance at scale. Great results so far in terms of uncovering critical performance issues and cutting call analysis time.

We are now evaluating the AI Voice Agent.