GenerativeAI One Year After ChatGPT: Where are we?

Steering well clear from a running commentary of OpenAI governance issues, today we reflect on the last 12 months as ChatGPT approaches its first "birthday".

Focus On: ChatGPT One Year After

AI, even transformative models, were not new one year ago but the release of ChatGPT was the Nokia 3210 moment for generative AI. Just as the Nokia 3210 put a mobile phone in everyone's hands, ChatGPT made the concept of Generative AI understandable by everyone and captured the imagination of billions of people, including CEOs.

Within the span of just a year, this technology has not only gained prominence but has become a cornerstone of innovation, especially within Fortune 500 companies where its adoption is redefining the paradigms of sales and marketing.

ChatGPT's introduction marked the beginning of a new chapter where traditionally manual and time-consuming tasks were reimagined. Fortune 500 firms have been at the forefront, embracing these newfound capabilities to bolster their competitive edge. In reflecting on the past year, let us count the ways in which ChatGPT and similar large language models (LLMs) have uniquely contributed to sales and marketing:

Enhanced Lead Generation: ChatGPT has been instrumental in developing sophisticated lead qualification systems. Fortune 500 companies can now leverage AI to analyse lead data, predict buying patterns, and personaliae outreach, resulting in higher conversion rates.

Streamlined Customer Service: AI-driven chatbots, powered by ChatGPT and similar models, have been deployed to handle customer inquiries. This not only improves response times but also allows human customer service representatives to focus on complex, high-value interactions.

Innovative Content Creation: Content strategists are using LLMs to generate a variety of marketing materials—from blog posts and whitepapers to social media content—tailored to specific audience segments, significantly reducing content development cycles.

Sales Training and Enablement: Fortune 500 companies have been utilising ChatGPT's capabilities to create interactive training modules and real-time sales coaching, ensuring that their teams are well-equipped to meet the evolving demands of the market.

Market Analysis and Intelligence: Generative AI has empowered marketers to perform rapid market trend analysis, providing insights that help shape product development and go-to-market strategies far more effectively than before.

Personalised Marketing at Scale: Through generative AI, enterprises are able to personalise customer experiences at a scale previously impractical. Be it hyper-relevant email marketing campaigns or website experiences, the ability to individualise customer interactions has significantly improved engagement and loyalty.

Automation of RFPs and Proposals: Sales teams have been able to automate significant parts of the RFP process, leveraging generative AI to create tailored proposals that resonate with prospects, reducing turnaround times, and improving win rates.

Sales Forecasting: By integrating generative AI models, companies have seen advancements in predictive sales analytics, leading to more accurate sales forecasting and strategic resource allocation.

Challenges Ahead

As mentioned in previous issues, I believe brands need to very careful about data provenance and ethical sourcing. The considerations are essential to protect the brand in the long term. Whilst less prudence is required when deploying internal use cases (see Sales Forecasting), when LLMs are used to generate content which will be delivered to end users, adherence to copyright law and protecting the brand audience from toxicity becomes key.

The availability of GPUs, used to build, train and run the models, is also an issue, especially for brands willing to embark on the journey of in-house model building. Whether an enterprise buys its own hardware or runs its models on a private cloud, they will quickly discover the cost of running generative AI and the need for vertical, focussed applications.

Finally, it is worth mentioning that the quest into finding the right balance between ethical and morally acceptable AI outputs and allowing the AI to have a mind of its own, free from our own biases, is proving quite challenging. Many maintain that ChatGPT output performance has degraded, becoming more "vanilla", because of the many ethical filters applied to it in the aftermath of the global reaction that followed its launch. I think poor user input/feedback is also to blame, still I predict we will wrestle with these issues for a while before striking the right balance.

Follow me

That's all for this week. To keep up with the latest in generative AI and its relevance to your digital transformation programs, follow me on LinkedIn or subscribe to this newsletter.

Disclaimer: The views and opinions expressed in Chronicles of Change and on my social media accounts are my own and do not necessarily reflect the official policy or position of S&P Global.