OpenAI now offers ChatGPT 4, or GPT-4, only through its Plus subscription service of $20 a month.
According to OpenAI, the upgrade yields better results: “In a casual conversation, the distinction between GPT-3.5 and GPT-4 can be subtle. The difference comes out when the complexity of the task reaches a sufficient threshold—GPT-4 is more reliable, creative, and able to handle much more nuanced instructions than GPT-3.5.” To document it, the website provides comparisons of GPT 3.5 and 4’s performance on standardized test.
Subscribers to ChatGPT Plus can now use it with plugins. OpenAI has a plugin to enable ChatGPT to connect to and draw from the Internet.
testing gpt-3.5 v. gpt-4 by asking the same question
We did our own comparison, asking GPT-3.5 and GPT-4 the same question (one we explored in a prior post) related to whether AI will cannibalize Google search’s ad revenues:
Question: Will ChatGPT integration into online search, including in Google's own search, cannibalize the total ad revenues of Google, given that people won't need to visit websites or click through to get the information they want?
The answers confirmed that GPT-4 is better–indeed, much better–than GPT-3.5.
gpt-3.5
As an AI language model, I don't have access to up-to-date information about specific business strategies or arrangements. However, I can provide some general insights on this topic. Integrating AI language models like ChatGPT into online search engines could potentially change the way people access information. By providing direct answers and information within the search results, users might not need to visit websites or click through as frequently If users find the information they need directly from search results, it's possible that some websites could see a decrease in traffic, which could impact their ad revenues. However, it's important to note that search engines like Google rely on a variety of factors to generate ad revenue, such as sponsored ads, pay-per-click campaigns, and other advertising models. Moreover, AI language models like ChatGPT can also enhance user engagement and improve the overall search experience. They can provide more accurate and relevant answers, leading to increased user satisfaction. This, in turn, can help search engines retain and attract more users, potentially leading to increased ad revenues in other areas. Ultimately, the impact of integrating AI language models into search engines on ad revenues would depend on various factors, including the implementation strategies, user behavior, and the effectiveness of the monetization models employed by search engine providers.
gpt-4
The integration of AI like ChatGPT into online search could theoretically affect Google's ad revenues, but it's important to consider the complexity of the situation. Firstly, it's important to remember that ad revenues are generated not just through clicks, but also through impressions. Even if users are not clicking through to websites as often, they would still be exposed to ads during their search and interaction with the AI. Secondly, it's not guaranteed that AI integration would entirely replace the need to visit websites. While an AI like ChatGPT can provide quick answers to simple questions, it may not always be able to provide the depth or breadth of information that can be found on a full webpage, particularly for complex topics or unique queries. In such cases, users may still need to click through to websites for more information. Additionally, Google could adapt their business model in response to these changes. For instance, they could develop new advertising products or strategies that are specifically designed to work with AI-powered search. They could also find ways to monetize the AI interactions themselves, such as sponsored answers or prioritized content. Ultimately, while AI integration could potentially affect Google's ad revenues, it's also likely that the company would adapt and innovate in response to these changes. Google has a history of staying ahead of the curve when it comes to technological advancements, and it's likely they would find ways to leverage AI integration to their advantage. Remember, however, that these are just possibilities and the actual impact would depend on a variety of factors, including the specifics of how AI is integrated into search, how users respond to these changes, and how Google and other companies adapt their strategies in response. It would be a complex interplay of technology, user behavior, and business strategy.
Review
We think GPT-4 provided a more informative answer. We put in bold some insights of GPT-4 that were not contained in the answer from GPT-3.5.
For example, GPT-4 noted that Google makes ad revenue through impressions, not just clicks. GPT 4 also challenged the premise of the question that chats will always be able to replace click throughs to websites, especially for more complex queries or the need for more detailed information. GPT-4’s answer was more nuanced and insightful, and provided a stronger analysis.
If you’re impressed already, just wait for GPT-5!