How much can AI systems like ChatGPT really help your organization?

In my position at NTT, I am always looking for innovative ways to improve our organization's efficiency and overall performance, and to deliver value to our clients.

In the past year, new AI systems released to the public – including ChatGPT, launched by OpenAI – have become a topic of much interest and debate. They offer remarkable generative capabilities, rooted in natural-language processing and trained on huge data sets, although there are meaningful limitations to keep in mind, as I will explain.

• ALSO READ → 5G and AI usher in a new era for Mobile World Congress

In this article, I wanted to explore how the integration of AI systems can contribute significantly to business growth. So I thought, why not demonstrate what I mean by asking ChatGPT why AI is becoming a valuable business tool? Here’s what it said, with a few small tweaks:

8 ways AI can make a difference in your organization

1. Boosting marketing and sales: AI systems can analyze vast amounts of data to provide valuable insights into customer preferences and behavior. Marketing and sales teams can use these insights to develop more targeted and effective strategies. ChatGPT and other AI systems can also help to create personalized content for customers, increasing engagement and conversion rates, or identify potential sales leads so that sales teams can focus on high-value prospects.
2. Improving product development: AI can play a significant role in this area. By analyzing customer feedback and market trends, these tools can help identify potential areas of improvement or new product opportunities. Furthermore, AI-powered tools can be used to optimize product design by simulating scenarios and analyzing the outcomes. By incorporating AI into product development, organizations can develop more innovative products that better cater to their customers’ needs.
3. Expediting decision-making: AI systems can analyze massive datasets to identify patterns, enabling organizations to make more informed decisions and gain valuable insights into customer behavior, market trends and potential risks. This information can be used to develop strategies and make data-driven decisions, ultimately leading to increased competitiveness and growth.
4. Optimizing marketing and sales efforts: By analyzing user behavior and preferences, AI-powered marketing tools can help enterprises understand their target audience better. This enables businesses to create personalized marketing campaigns and product recommendations, resulting in higher customer engagement and conversion rates. Moreover, AI systems can also automate repetitive sales tasks, such as lead-generation and lead-scoring, allowing sales teams to focus on more critical aspects of their work.
5. Enhancing internal communications: AI systems can automatically generate meeting agendas, minutes and summaries, saving time and ensuring consistency in documentation. Moreover, AI-powered virtual assistants can help employees manage their schedules, set reminders and even provide instant answers to work-related queries. This leads to better organization and increased productivity across the enterprise.

• ALSO READ → Invest in AI and your employees for CX success

6. Enhancing security measures: Cybersecurity is a concern for all organizations as cyberthreats become more sophisticated. AI systems can identify vulnerabilities and potential threats by analyzing network traffic, user behavior and other data. In this way, organizations can strengthen their security posture and protect their valuable assets from cyberattacks.

• ALSO READ → How AI keeps the wolves at bay in cybersecurity

7. Automating routine tasks: AI systems can automate a range of routine tasks, from data entry to invoice processing, which can save time and resources. By automating these types of repetitive tasks, organizations can enhance efficiency, limit the room for human error and allow employees to focus on more strategic and innovative tasks. This not only improves productivity but also fosters a culture of innovation within the organization.
8. Streamlining customer support: One of the most significant applications of AI systems like ChatGPT is in customer support. By implementing AI chatbots, organizations can provide 24x7 customer support without the need for human intervention. Chatbots can handle multiple customer queries simultaneously, ensuring prompt response times and reducing the workload on human customer-support staff. This results in improved customer satisfaction, a better brand image and increased operational efficiency.

Why we still need humans

ChatGPT’s answer to my query was a good start to my writing process, and it demonstrates just some of the benefits listed here. It would have taken me more than a few seconds to compile this list, from thinking about what I wanted to say to structuring the piece and then putting it all together. So, it was a great time-saver in giving me key points and an outline.

But there’s a reason you’re reading this blog, not an AI-driven answer to this question, and I’m guessing it’s because you’re looking for an informed, human perspective. Here goes ...

What this exercise brought home to me was that generative AI doesn’t know anything in the conventional sense – it’s just amazingly good at predicting sequences of words relevant to specific contexts, based on the enormous data sets provided by its creators (in the case of ChatGPT, these data sets are also a couple of years old). So, it’s essential to fact-check the answers these tools provide because they tend to “hallucinate” – that is, invent facts that sound remarkably realistic.

You’ll also notice that this blog includes links to articles NTT has produced on similar topics that reflect the topics and technologies we’re focusing on, based on our business strategy, current market research and what our clients are telling us they’re interested in. But if I’d relied on the ChatGPT answers for these, you’d have links to some outdated press releases instead – even some from other companies rather than ours. Not all AI tools use data that “stops” at a certain point in time, but you still need an intelligent human to frame their output in the right context.

Continuing with our experiment, ChatGPT also has no real concept of ethical, legal and compliance considerations, and this is still a gray area in so many ways: who gets the blame when an AI tool makes a costly mistake, breaches data privacy guidelines or demonstrates blatant or hidden bias, for example?

Team up with an expert to make the most of AI

ChatGPT is just one example of an evolving range of AI tools with myriad applications, each with its own set of limitations. To get the most out of them, I recommend having access to expert humans who set up these systems, integrate them with other parts of your organization, and train and manage them from day to day in a way that’s aligned with your business strategy. Then, they can certainly “play a significant role”, as ChatGPT said.

Despite the apparent user-friendliness and “wow factor” of many of these tools, making them work optimally in a business environment can quickly become complicated and expensive. This is where there is value in working with a managed service provider like NTT that already has deep experience in integrating AI into operations (think AIOps), customer support, infrastructure management and many other business areas.

We already use OpenAI’s latest GPT-4 model and ChatGPT to gather external, publicly available data insights. We combine these with our internal, enterprise-grade data analysis tool that we provide in collaboration with SparkBeyond, a technology provider focusing strongly on the application of AI in strategy, business growth and innovation.

In doing this, we give our clients the best of both worlds: visibility of trends based on publicly available data as well as granular insights based on their internal data (to which models like ChatGPT are blind) into the performance of their organization.

We also help them make sense of the data and use it to achieve their objectives – which we can do only once we have a deep understanding of their industries, markets, customers and goals. And for this, we rely on human connections, conversations and intelligence.