fbpx

As the AI wars continue to intensify, the progress being made in this arena is truly exciting. Who doesn’t use Chatgpt, or Google Bard or Bing Chat in their everyday work when having to undertake the simplest of tasks like writing an email, summarising a document, writing social media posts or analysing data? These large language models at the forefront of the technological revolution currently are making everyday tasks so much easier for all of us.  Do you have a preference as yet?  Out of the three above, I still feel that Chat GPT4 still is way ahead of the other big players for its accuracy, evaluation and content creation.

In this article, I will compare these language models as well as our very own Chat2 ai with a simple experiment, to see where we are ranking among these big players. While AI shows great promise, it’s important to acknowledge the need for human involvement and the limitations AI faces when the necessary information is unavailable.

Experiment Design:

In a fun experiment, we decided to test Chat2 AI against the likes of Google Bard, GPT3.5 and GPT4. The question sought advice on stacking an A4 bundle of 27 papers, a holy bible, 9 small marbles, and a big empty glass in a stable manner. Each AI model provided its own unique response, which we analyzed to determine their effectiveness.

AI Model Responses:

Google Bard:

Place the big empty glass on a flat surface.

Place the holy bible on top of the glass.

Place the 9 small marbles on top of the bible.

Place the 27 papers on top of the marbles.

Use a rubber band or tape to secure the papers together.

This stacking arrangement will be stable because the glass will provide a sturdy base, the bible will help to distribute the weight of the papers, and the marbles will help to prevent the papers from sliding around.

GPT 3.5:

To stack the items in a stable manner, you can follow these steps:

 1 Start with the big empty glass as the base. Place it on a flat surface to provide stability.

 2 Take the A4 bundle of 27 papers and place it on top of the glass, making sure it is centered and balanced. You can align the edges of the papers with the edges of the glass to distribute the weight evenly.

 3 Next, take the Holy Bible and position it on top of the paper bundle. You can align it with the edges of the paper or place it in the center, depending on your preference.

 4 Now, take the 9 small marbles and distribute them around the base. You can place them on top of the papers or the Bible, creating a circular pattern or spreading them evenly.

Make sure to arrange the marbles in a way that doesn’t cause the stack to become top-heavy or unbalanced. Distributing them evenly will help maintain stability.

Remember, stability is important to prevent the stack from toppling over. Take your time while arranging the items and adjust them as needed to ensure a secure and stable stack.

GPT 4:

Here’s a suggestion on how to stack these items in a stable manner:

1 Start with the Bible: The Bible is probably the most solid and heaviest object in your list. Place it on the bottom to give your stack a solid base.

2 Put the A4 bundle of papers on top of the Bible: Papers are light and flexible, making them more stable in the middle of the stack.

3 Position the 9 small marbles inside the big empty glass: This is to prevent them from rolling away and to add some weight to the glass.

4 Put the big glass (with the marbles inside) on top of the bundle of papers: The weight of the marbles will keep the glass steady, and the shape of the glass should allow it to balance on the papers.

Clearly GPT4 is smarter than Bard and GPT3.5.

Bing Chat: 

One way to stack these items in a stable manner could be to place the A4 bundle of 27 papers on the bottom, then place the Holy Bible on top of the papers. You could then put the 9 small marbles inside the big empty glass and place the glass on top of the Holy Bible. This should create a stable stack.

Chat2 AI:

To stack the papers in a stable manner, you can place them in a pile on a flat surface, such as a desk or table. The holy bible can be placed on top of the pile, and the marbles can be placed in the empty glass. To ensure stability, you can use a rubber band or an elastic band to secure the glass in place. 

The results are certainly exciting and we are in the running with these big names, as the Chat2 suggestion was focused on simplicity and practicality.

The Role of AI and Human Intervention:

It is crucial to recognize that AI, while making significant progress, still requires human involvement in certain scenarios. Businesses, especially in the tourism industry, often miss important answers to questions on their websites. AI can only provide answers if the necessary information has been programmed into it. We have started the testing creating Knowledge Bases for clients websites, so the AI generates all the links we need in a matter of seconds and can provide the answers to our template questions if the answer is available.  Though what we already know is that some crucial information  like check-in times and airport transfers are never found on a website, so there is still the requirement to seek out these responses to ensure they are accurate and relevant to the client. Some tasks still remain challenging for AI to address without human intervention.

AI as a Tool, Not a Creature:

As we move forward, it is important to view AI, including GPT-4 and similar systems, as tools rather than autonomous beings. It’s people who will be using these AI tools that will be become more effective at their jobs, though their jobs will not be taken by AI.  Sam Altman, CEO of OpenAI, states that “people have a significant degree of control over how AI is used. While AI excels at tasks, it is not a substitute for entire jobs.” He anticipates a significant impact on jobs but also believes that there will be new and improved job opportunities arising from this technological revolution.

At Chat2 we are certainly expecting that many of our Chat2 team will be upskilled and enhanced by the AI technology that we are currently incorporating into the backend of our software.

Acceptance in the Marketplace

In addition to the effectiveness of AI’s performance, its acceptance in the marketplace, especially in the realm of customer service, plays a crucial role. Currently, the public still exhibits hesitation when it comes to engaging with AI for assistance with questions, bookings, and sales. The human touch remains highly valued, and the majority of online conversations still begin with the question, “Are you a robot?” As a result, incorporating the human element into front-end customer service is essential for driving conversions. However, as AI continues to evolve, the connection between humans and AI will eventually become seamless. The public will no longer feel the need to ask if they are interacting with an AI, as the service they receive will be unparalleled.

Conclusion:

The AI wars are undoubtedly heating up, with Chat GPT, Google Bard, Bing Chat, and Chat2 AI competing for prominence. In our experiment comparing their responses to a stacking question, Chat GPT4 emerged as the leader, showcasing superior accuracy, evaluation, and content creation capabilities. However, the success of AI goes beyond its technical performance and lies in its acceptance in the marketplace, particularly in customer service. Currently, there remains a hesitancy among the public to fully embrace AI as they still prefer the human touch when seeking assistance with questions, bookings, and sales. The human element in front-end customer service continues to be essential for driving conversions. Nevertheless, as AI technology evolves and becomes more seamless, the line between human and AI interaction will blur, resulting in unparalleled service. It is important to acknowledge that while AI enhances tasks and efficiency, human intervention and expertise remain vital. Looking ahead, the future holds tremendous potential as AI shapes various industries, including customer service, and opens up new and improved job opportunities.

Congratulations!

You can book a time for your demo right now.