We’ve all heard the recent hype about a new artificial intelligence (AI) technology, popularly known as generative AI, dominating the headlines.
There are many parallels between the explosion of AI and the concept of digital transformation that was popular more than 15 years ago. Companies have realized that every part of their business needs to be enabled by technology to stay competitive and vibrant. AI is the logical next step in the age of digital transformation.
It can seem daunting for a director to make the learning curve and decipher what is hype and what is a viable business opportunity.
Given the impact of this technology and its potential disruption to normal business, it is time for boards to think about it.
Here is a general overview to help the board community understand these exciting new developments and how they can be applied to different business functions.
Generative AI: a A type of artificial intelligence that can develop new content based on data rather than just recognizing and classifying. Generative AI is multimodal, using text to create new outputs such as text, images, audio, video, code, and simulations. It does this by learning from large datasets of existing content. For example, a generative AI model trained on a dataset of texts can be used to create new texts such as poems or stories. Generative AI systems are broadly classified as a subset of machine learning. Various generative AI systems are currently available in the market at various stages of maturity, such as ChatGPT and Bard. Generative AI is becoming increasingly popular because it can be used to develop realistic and engaging content.
Large Language Model (LLM): A type of artificial intelligence (AI) model trained on large datasets of text and code. With LLM, you can generate text, translate languages, create different types of creative content, and answer questions in an informed way. LLMs are trained using a process called deep learning. Deep learning is a subset of machine learning that uses artificial neural networks to learn from data. Artificial neural networks are inspired by the human brain and can learn complex patterns from data. LLMs are trained on large datasets of text and code. These datasets can contain billions of words and cover a wide range of topics. This allows LLM to learn statistical relationships between words and concepts.
ChatGPT and Bard: ChatGPT and Bard are both well-known generative AI chatbots in mainstream media. ChatGPT is developed by OpenAI and Bard is developed by GoogleGOOG AI. These chatbots have been trained on large datasets of text and code and can be used to generate text, translate languages, create different types of content, and answer questions in informative ways.
What’s the difference?
There are some key differences between these two popular chatbots. At a basic technical level, Microsoft’s MSFT ChatGPT and Google’s Bard employ different large-scale language models. ChatGPT uses Generative Pre-trained Transformer 4 (GPT-4), while Google BARD relies on a custom language model for Dialogue Applications (LaMDA).
One important difference that is particularly relevant to everyday users is that Bard can access the Internet. Bard can pull responses from the Internet in real time. ChatGPT relies on datasets that only pass through 2021. This is a limitation that may prevent ChatGPT from providing real-time updates.
One particularly differentiating aspect of Google’s Bard is its end-to-end sovereignty over data, training, models, etc., which remain private to the business, allowing complete control over their IP. .
Customer Use Cases for Generative AI
- Content Creation: Can be used to create new content such as social media posts.
- Customer Service: Can be used to answer customer questions and provide support.
- Sales and Marketing: Can be used to generate leads, identify prospects, and close deals.
- Research and Development: Can be used to generate new ideas, explore new markets, and develop new products.
Notes
Potential drawbacks of generative AI include:
- Bias: Generative AI models are trained based on data. If the data is biased, the model will be biased as well. This can cause the model to produce discriminatory or unfair output.
- Lack of control: Generative AI models can be difficult to control. When a model is trained, it may produce different output than what the user intended. This can be a problem if the model is used to generate sensitive content such as medical or financial information.
- Cost: Developing and maintaining a generative AI model can be expensive. This can be a barrier for small businesses and organizations with limited budgets.
A drawback that many everyday users are familiar with is the potential for incorrect information to be included. Generative AI models can be used to create surreal fake content known as “deep fakes”. This can be used to spread false information or damage someone’s reputation or the company’s reputation.
One well-known example of this is a recent photo of Pope Francis wearing a stylish white down jacket and a bejeweled crucifix. This image was created with the generative AI software Midjourney. The photo garnered millions of views and many social media users felt cheated. This is a relatively harmless example of how malicious actors can use these tools to spread misinformation.
As the technology landscape evolves more rapidly than ever before, directors must be prepared to make thoughtful business decisions and provide informed oversight.
I believe it is good for directors to have a learning curve and immerse themselves in the world of generative AI. As directors, we all know companies need to act quickly to take advantage of these unprecedented opportunities. At the same time, as a board member, I need to make sure that the risks are contained. This is never easy, but it is more important than ever in this new age of AI.
Early adopters of technology that improve customer experience and enable different aspects of their business tend to be the winners. Those who are slow to innovate will not succeed.
The rapid advances in AI deserve serious consideration and study of how it can be applied to specific industries. Perhaps start integrating with small use cases that are “easy to win”. As this technology continues to evolve and improve, it is important for businesses to better understand its capabilities and future potential.
The board I work on already invites outside experts (strategic consultants like Bain and AccentureACN, or major vendors like Google) to the board room to provide tutorials and overviews of generative AI and how it identifies companies. shares how it relates to the industry of
follow me twitter Or LinkedIn. check out You can find my website and other works here.
