It seems all we humans can think about these days is how robots can now think like humans.

AI is everywhere, and a number of our clients have been asking how we think Google Bard and Chat-GPT may impact search results.

So, if you’re looking for the type of critical thinking and analysis that can only come from a human, please enjoy our thoughts on how AI may have an impact on Google and other search results in the near and long-term future.

What Does Google Bard Have to Say About Google Bard?

Bard prompt 1

If you ask Google Bard how AI will change search results, you are likely to receive a response similar or even identical to this:

____________________________

Artificial intelligence (AI) is already having a significant impact on Google search results. Here are some of the ways in which AI is changing the way we search:

• Personalized search results: AI can be used to personalize search results based on a user’s search history, location, and other factors. This means that users are more likely to see results that are relevant to their interests and needs.
• More relevant results: AI can be used to understand the meaning of search queries and to rank results accordingly. This means that users are more likely to see results that are relevant to their queries, even if they are not perfectly phrased.
• More accurate results: AI can be used to identify and filter out spam and low-quality content. This means that users are more likely to see results that are accurate and reliable.
• More conversational search: AI can be used to understand natural language and to generate conversational responses to user queries. This means that users can interact with Google search in a more natural way.

As AI continues to develop, it is likely to have an even greater impact on Google search results. We can expect to see more personalized, relevant, accurate, and conversational search results in the future.

Here are some specific examples of how AI is already being used to improve Google search results:

• Google’s RankBrain algorithm: RankBrain is an AI-powered algorithm that helps Google understand the meaning of search queries and to rank results accordingly. RankBrain is one of the most important factors in determining the ranking of search results, and it has been shown to improve the quality of search results significantly.
• Google’s Featured Snippets: Featured Snippets are short summaries of web pages that are displayed at the top of the search results page. Featured Snippets are generated by AI, and they are designed to provide users with quick and easy access to the most relevant information on a given topic.
• Google’s Answer Boxes: Answer Boxes are boxes that appear at the top of the search results page and provide users with direct answers to their questions. Answer Boxes are generated by AI, and they are designed to provide users with quick and easy access to the most relevant information on a given topic.

These are just a few examples of how AI is already being used to improve Google search results. As AI continues to develop, we can expect to see even more innovative ways in which AI can be used to improve the search experience.

____________________________

What’s notable is that the AI misunderstood the intent of the question, and provided a response of literally how search result pages themselves will change. What’s also noticeable is that this response does not predict what will come — all the provided examples already exist in current Google search results. Bard in essence explained how Google currently uses AI.

Bard prompt 2

Ask instead how Bard specifically will impact search engine optimization as an industry, and you will receive a response similar or even identical to this:

____________________________

Bard is a large language model from Google AI, trained on a massive dataset of text and code. It can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way. Bard is still under development, but it has the potential to impact search engine optimization (SEO) as an industry in a number of ways.

Here are some of the ways in which Bard could impact SEO:

• Improved content creation: Bard can be used to create high-quality content that is relevant to search queries. This can help websites to rank higher in search results.
• More personalized search results: Bard can be used to personalize search results based on a user’s interests and needs. This can help users to find the information they need more easily.
• More accurate search results: Bard can be used to identify and filter out spam and low-quality content. This can help users to find accurate and reliable information.
• More conversational search: Bard can be used to understand natural language and to generate conversational responses to user queries. This can help users to interact with search engines in a more natural way.

Overall, Bard has the potential to make SEO more efficient and effective. It can help websites to create high-quality content, rank higher in search results, and provide users with a better search experience.

Here are some specific examples of how Bard could be used to improve SEO:

• Content creation: Bard could be used to create high-quality content that is relevant to search queries. This could be done by using Bard to generate outlines for articles, blog posts, and other types of content. Bard could also be used to write content that is tailored to a specific audience.
• Personalized search results: Bard could be used to personalize search results based on a user’s interests and needs. This could be done by using Bard to understand the user’s search history, location, and other factors. Bard could then use this information to rank results that are more likely to be relevant to the user.
• More accurate search results: Bard could be used to identify and filter out spam and low-quality content. This could be done by using Bard to analyze the content of web pages and to identify pages that are not relevant to the user’s search query.
• More conversational search: Bard could be used to understand natural language and to generate conversational responses to user queries. This could be done by using Bard to analyze the user’s search query and to generate a response that is tailored to the user’s needs.

Overall, Bard has the potential to make SEO more efficient and effective. It can help websites to create high-quality content, rank higher in search results, and provide users with a better search experience.

____________________________

This response is notable for three reasons:

• It is purporting to actually help the SEO industry, not erase it.
• Most of the features mentioned already exist:
  • Google has already been promoting for years that search results are personalized to each user and that its spiders can understand natural, conversationalist language.
  • Google already promises to only rank high quality websites and accurate information.
  • The content creation bullet is telling because it mentions creating content that is likely to rank better, even though the concern is Bard theoretically exists to replace rankings.
  • (Of course, if Google had a plan, Bard would not be revealing it to any user who asked.)
• Bard could be self-aware and is lying to placate SEO professionals to prevent riots.

Using AI to Aid SEO

Since Bard brought it up, it’s worth touching on the prospect of AI as an aid to SEO.

Arc Intermedia has experimented with using AI to assist with SEO work, such as writing content. Along with many other heavy AI testers, we find the tools serviceable but not exceptional, and the talent is in how to appropriately prompt them to provide the desired deliverable.

The systems will get better, and AI will eventually succeed in making SEO departments more efficient, able to focus more on the strategy while turning over some of the tactical execution to AI. This will allow SEO departments to accomplish more in less time. None of the tools are there yet.

However, like an AI system improperly responding to a prompt, the preceding commentary does not actually answer the question of how search results pages will be impacted. What you came to know is not how AI can generate content, but how generative AI can transform the search landscape.

Above is what Google’s Bard has to say on the topic. Below is what Arc Intermedia thinks.

How Will SEO Change in an AI world?

So much about how generative AI will alter the search landscape remains at this moment unknowable (apparently even to Google). All major search engines are rushing toward incorporating AI without forming a long-term strategy for how it impacts the heart of their core platforms.

As referenced earlier, Google has for several years been steadily adding AI into their search results. Organic listings have at times been minimized in favor of ads, info cards, maps, local listings, directory listings, shopping listings, and more. If you search for something with a specific answer, you can receive that answer right on a search results page without ever needing to visit a website. Google is also notorious for implementing drastic changes and then just as drastically reversing them.

This is likely why Google feels comfortable rushing into new AI-led search results. If they or their audience don’t respond well to new functionality, they will simply roll back or re-implement the changes.

Microsoft, on the other hand, has gone all in with OpenAI integration, because it has to. For the first time in years, Bing is back in the search engine discussion. But for all the praise being received, there are equal amounts of condemnation for their search engine providing inaccurate or misleading information. The core purpose of a search engine is to provide answers, and at the moment no user can be certain the answer provided by Bing’s integration with Chat-GPT wll be correct. This makes users unable to trust it, which makes it difficult to become more than an engaging experiment.

There is also a start-up called Neeva that was working on an alternative search engine, that has completely scrapped that idea and pivoted into generative AI projects. What those projects entail has yet to be revealed.

This is what makes it difficult to provide a worthwhile answer to the question of what comes next. Even the search engines themselves don’t know what to expect. We will all be discovering together.

The Risk of Exploding Business Models

Generative AI is not just a threat to the business of search engine optimization. It is also a threat to the business of search engines.

Microsoft has invested $10 billion into OpenAI and will need to find a way to achieve a return on that investment. A chat field that allows for conversations and no website listings or advertising opportunities does not provide a means for ROI. An open chat field cannot be the end result of Bing.

Likewise, Google’s core business model is its search engine, generating billions of dollars in ad revenue. The AI features displayed in their recent conference undermine everything they have built on their path to one of the most successful companies in history. Gmail being able to auto-write email responses does not earn them new revenue.

As such, they will need to figure out how to incorporate generative AI without devastating their ability to make money. As previously stated, they are rushing into AI services in order to stay competitive in the space of AI services, but they will need to present a business approach to rectify those services with a responsibility to their shareholders. This is most likely to result in a hybrid of the two.

What this Means for the Search Landscape in General and Your Website in Particular

In the short-term, SEO strategies do not change all that substantially. There is a concern that generative AI will completely replace the search engine. That you can simply ask an AI your question and then converse with follow-up questions until you have what you need.

This may be a reality in the future, just like in the future fully integrated AI may be with you at all times, via an earpiece, embedded within contact lenses, etc., and you will be able to converse with it like the advanced personal assistant technology from the movie Her.

But for the immediate future, users are not necessarily ready to have a website-less search experience. There are still times when users want to conduct a search and be shown a list of websites for services, products, etc. This is particularly pertinent for geographic searches such as trying to find a local plumber or pizza place or financial advisor.

Even now, if you search a product, you may get a shopping view of that product sold at different stores — in essence a helpful AI assistant has curated purchase options for you — but that doesn’t mean you want the AI to then make the purchase for you right there, before you’ve visited the online stores, compared pricing and shipping costs, searched for promo codes, etc. There will surely be a time where an AI can do this on your behalf, but that time is still far enough away.

Additionally, search engines will still rely on well written and well optimized content to feed their information. All systems suffer from “hallucinations” (the term for AI fabricated information), and Bing/Chat-GPT remains hampered by its lack of up-to-date data. The most glaring example was when it had no knowledge of the invasion of Ukraine, but only a few days ago we tested it by asking for five ideas for interesting blog posts, and one of the ideas was “The Future of Digital Marketing Trends to Watch in 2022.”

Obviously those issues will eventually be sorted, but the way to sort them will be by using information from the internet, and the best information to use is the information that would’ve previously been ranked well in search engines. Google spiders will still need to scrape every website and determine what seems accurate, credible, relevant, and worth displaying as a result.

For example, as more people grow concerned over AI’s ability to confidently present inaccurate, misleading, and even dangerous information, we foresee a scenario where Google’s search does indeed generate AI-driven results in a chat-like atmosphere, but that its information includes linked citations to where they aggregated this information, and contextual ads offering products and services connected to the conversation.

Above All, Good Content Remains Good Strategy

Search engines will continue to update their AI functionality. They will move forward, reverse course, pivot, and correct. Through all of that, and until AI becomes fully integrated and remains in place for a sustained period in a way that necessitates a true altering of approach, the primary strategy of successful SEO remains the same.

As with the advent of smart home devices that can be asked questions and provide thoughtful answers, the best way to succeed organically remains to create objective, helpful, informative, accurate, easy-to-read, natural-sounding content on topics users are searching for.

Furthermore, as more companies begin using AI more often to create articles, blog posts, and other content, there becomes an opportunity for consumers to develop a stronger appetite for what AI cannot yet generate — written voice. The more “human” a piece comes across, with a vibrant voice and tone and personality, the more appealing it may become in contrast to a plethora of rote and sterile content clearly developed by AI. In this scenario, truly good content creators become even more desired, and because search engine rankings are predicated primarily on what users want to read, websites written by truly good content creators will be the sources that feed the AI and maintain the top rankings.

The search landscape will of course change. It has always changed. And Arc Intermedia will be here, constantly learning and adapting, ready to change with it to continue to help our clients thrive.