The integration of artificial intelligence (AI) into search engines has been an ongoing process since the early days of SEO. However, recent events have marked a significant shift in the industry. In late 2022, ChatGPT, an AI-powered search chatbot backed by Microsoft, was launched, kicking off the race for AI search chatbots.
In this article, we will explore the latest developments in SEO and AI, including how AI is being used in search engines other than Google, and what this means for the future of online marketing. You can also watch our team discussion on the topic below.
Table of Contents
Summary of Events so Far
- November 30th, 2022 – Microsoft-backed ChatGPT was launched. It was developed by OpenAI, the brains behind AI image generator DALL-E, and signified the start of the AI search chatbot race.
- February 6th, 2023 – Google Bard was announced with very limited release to “trusted testers.”
- February 7th, 2023 – Microsoft’s OpenAI partnership enabled the surprise launch of AI-powered Bing on the 7th of February 2023. Users who complied with requests to download Edge and set Bing as their default search engine could move up the waitlist and start playing. This beat Google to the punch with AI-assisted search in the public domain.
- February 8th, 2023 – Alphabet, Google’s parent company, experienced a hefty drop in share value due to a botched Google Bard press release and a demo that returned incorrect information.
- February 9th, 2023 – Bing AI was demo’d at a press conference, and similarly served incorrect responses. Interestingly, the focus was still on Google’s mistake, and Bing’s largely went unnoticed.
- March 14th, 2023 – The highly anticipated GPT-4 was released, with better accuracy, creativity and the ability to respond with up to 25,000 words. It is also capable of accepting visual inputs, but this is still being tested.
- March 16th, 2023 – Microsoft takes its 365 suite to the next level with Copilot. Copilot seamlessly integrates with Microsoft 365, working alongside Word, Excel, PowerPoint, Outlook, Teams, and more, boosting creativity, productivity, and skills. There’s also a feature called Business Chat, which combines the power of LLM, Microsoft 365 apps, and your personal data including calendar, emails, chats, documents, meetings, and contacts.
- March 23rd, 2023 – OpenAI released plugins to allow ChatGPT to access the internet. No more working with data from 2 years ago! The process uses Bing’s search API to browse the web and handily includes sources within replies. Initially only available to ChatGPT Plus subscribers and selected developers.
- May 10th, 2023 – Google announces a whole new host of AI features and Bard additions. The biggest of which was SGE or Search Generative Experience. Still in the experimental stages, it has caused palpable shock waves within the search community. Find out more about the release and what we know about SGE so far.
- May 23rd, 2023 – At Build 2023, Bing announced it was expanding Copilot to the Windows 11 OS. Rolling out from June 2023, Windows Copilot aims to be your personal assistant right within your computer, helping run commands, generate content, retrieve information and more.
How Chat-GPT Changed Everything
In just a few months, ChatGPT acquired 100 million users, a feat that took Instagram two and a half years to achieve and TikTok nine months. Incredible. Needless to say, Google Bard is also expected to reach one billion users quickly once it is fully rolled out, due to the already high global usage of Google Search.
Above, we can see how long it took ChatGPT to reach its initial million users, in comparison to some of the biggest players in consumer applications. Source: AFR.com
As per Google Trends worldwide data, we can see a noticeable spike in interest in AI-related topics coinciding with the emergence of ChatGPT.
This surge in interest suggests that ChatGPT played a significant role in triggering the AI renaissance. The tool has undoubtedly contributed to the democratization of AI technology by making natural language processing more accessible to developers and businesses, spurring innovation and development in the field. The widespread adoption and popularity of ChatGPT have undoubtedly helped propel the world of AI forward, making it more accessible to people and businesses across the globe.
“AI has just changed the game by increasing the value I can provide as an SEO. From summarizing highly technical content to helping me brainstorm potential topics for obscure niches. // ChatGPT is only as good as the prompt it’s given, which lends itself as an invaluable tool for SEOs who are experienced working with algorithms, AI and ML through Google.” – Helene Jelenc from Flow SEO
Additionally, Ahrefs shows that between November 2022 and February 2023, roughly 800,000 articles were published on artificial intelligence (AI). This figure is likely much higher, emphasizing the complete saturation of the AI topic not just in SEO and marketing, but in general media, too.
The AI Hype Cycle
In September 2022, Gartner published a piece on the state of artificial intelligence and commented on its prediction of the developing hype cycle. Interestingly, this was before even ChatGPT was announced.
Gartner’s 2022 Hype Cycle for Artificial Intelligence (AI) highlights innovations in AI technology, expectations and rough timelines of when the so-called “plateau of productivity,” or essentially when consumers and businesses will be able to make the most use of the tech and integrate it into our everyday lives — when AI becomes “normal.” The report identifies four categories of AI innovations — data-centric AI, model-centric AI, applications-centric AI and human-centric AI.
Data and analytics leaders can use the Hype Cycle outlook to develop their AI strategies and prioritize technologies with high impact in the present. The report emphasizes the need to pay attention to innovations that will reach mainstream adoption in two to five years, such as composite AI, decision intelligence and edge AI.
Responsible AI and digital ethics in particular are important considerations for businesses in the adoption of AI technology into their workflows, products and tech stacks.
In seemingly no time at all, the web has been updated with new AI tools, products and resources — https://theresanaiforthat.com/ is a great source to check for the latest AI releases.
The Age of the Answer Engine
As if you needed any more abbreviations in your life, here’s a new one — AEO or Answer Engine Optimization. Although the term has been around since voice search entered the public domain, it’s found a new meaning and importance, with marketers scrambling to understand how they can optimize for Bing’s AI chatbot et al.
What is Answer Engine Optimization (AEO)?
Answer Engine Optimization (AEO) is the practice of optimizing content to increase its chances of being featured as a direct answer in search engine results pages (SERPs), often in the form of a featured snippet, knowledge panel and now in an AI search assistant. The goal is to provide the most relevant and accurate answer to the user’s query and have it selected as the answer shown by search engines.
Innovations and features like voice search technology, knowledge panels, answer boxes, and question hubs have empowered Google to deliver direct answers to users without requiring them to click on websites. This requires SEOs to think less about traffic and more about visibility and the sharing of voice.
AEO has evolved as Google has transitioned from a keyword-based search engine to an AI-driven platform that uses machine learning and natural language processing to analyze user queries and serve the most relevant — or correct — answer.
And now, with AI search chatbots, the question is how to optimize websites for answers provided in a conversational context. Check out the “How can I optimize for Google Bard, Bing AI and other answer engines” sections below for more on AE optimization techniques.
Is AI new to Search?
Not at all. In fact, AI is already an integral part of search engine technology, helping users find the most relevant and accurate results possible. Google has used artificial intelligence in its algorithms since 2015, when it rocked the SEO industry with RankBrain.
Google’s AI Algorithms
RankBrain – Announced Oct 2015
RankBrain is an artificial intelligence algorithm that Google uses to sort and understand search results. Unlike the old algorithm, which was fully hand-coded, RankBrain tweaks the algorithm based on the keyword, and then monitors how Google searchers interact with the new search results. It’s been shown to outperform human engineers in identifying the best page for a given search by 10%.
Image: RankBrain as depicted using DALL-E – prompted by Laura James
BERT – Announced Nov 2018
BERT (Bidirectional Encoder Representations from Transformers) is an AI technology used for natural language processing (NLP), natural language understanding (NLU) and sentiment analysis, which allows the algorithm to understand the meaning and intent behind a search query. Unlike previous Google search algorithms that processed words one-by-one in order, BERT processes every word in a sentence, taking into account the entire context of the search. This results in more accurate searches that better reflect the user’s intent, even accounting for prepositions that may change the context of the search.
MUM – Announced May 2021
MUM (Multitask Unified Model) is a new type of AI technology developed by Google to better understand complex needs in search. MUM uses the T5 model, a more powerful version of NLP transfer learning, to reframe NLP tasks and improve understanding of information.
Google has suggested that MUM could be used for document summarization, question answering, sentiment analysis and classification tasks. While it is not currently a ranking factor, Google has hinted that it could be in the future. MUM has been tested on vaccine information, personal crisis searches and improving spam protections.
LaMDA – Announced May 2021
LaMDA is a special language model that can chat naturally about a wide range of topics. What sets it apart is that it was trained on dialogue, not just text, which means it understands the subtle differences between open-ended conversations and other types of language. Its responses are evaluated on criteria like whether they make sense, are specific, interesting and based on facts. Google is also taking steps to make sure that machine learning models like LaMDA don’t perpetuate unfair biases or spread misinformation by creating a set of safety objectives called the Safety metric. The objectives train the model to avoid producing harmful or inappropriate content, such as violent or gory content, slurs or profanity. This is still a work in progress, and there is more to be done to improve the Safety metric.
How Search Engines other than Google use AI
Search engines other than Google also use AI in their algorithms in a variety of similar ways. For example, Bing’s AI capabilities allow it to understand the context behind search queries and provide more relevant results. Bing also uses AI to personalize search results based on user behavior and preferences.
Similarly, privacy-focused DuckDuckGo uses AI to understand the meaning behind search queries and improve the accuracy of results. Yandex, the most popular in Russia, uses AI to improve its image search capabilities, allowing users to find similar images more easily. Additionally, Baidu, the leading search engine in China, uses AI to provide voice search, natural language processing and image recognition capabilities. Since 2019, Baidu has been developing ERNIE, its own answer to ChatGPT, but there is still no sign of a public release.
Let’s be honest, Bing has been the plucky underdog of the search world for some time now. Seen as being kept alive by B2B Microsoft users and boomers who don’t know how to switch to Google, it’s been struggling to stay relevant.
However, on the 7th of February 2023, Microsoft revealed a partnership with OpenAI and announced AI-powered Bing to the world. Since then, the search engine has seen a marked resurgence in popularity. The Bing app is now the 10th most popular free tool app on the U.S. App Store, and Microsoft’s Edge browser is the third best Utility app. Almost overnight app intelligence firm, data.ai, estimated that the Bing app had seen a tenfold increase in new downloads globally.
More recently, Bing published a one-month retro on the launch of its AI chatbot, revealing that it had achieved 100 million daily active users — and counting.
Should I put more resources into Bing optimization?
It would be worth taking another look at your Bing assets — Bing Places and Bing Webmaster Tools — ensuring that you’ve got the basics covered and optimized whilst keeping the focus on Google. As Bing themselves said, “…we are fully aware we remain a small, low, single-digit share player.”
Google’s market share still far outweighs Bing, but the race to a fully-fledged AI search experience could see the needle move. It’s never too late to start optimizing for Bing, so take the time to review and explore the benefits that Bing can offer your company.
How Can I Optimize for Google Bard, Bing AI, and Other Answer Engines?
Honestly, if you’re regularly creating great quality, original, authoritative content for your readers or potential customers, you’re doing it.
“AI has the power to make every good SEO or content marketer better and every bad one worse. Don’t fall for the allure of 100% AI-generated content — use it to improve productivity, ideation, QA and everything in between.” – Laura James – Flow SEO
From what we know so far, we’d recommend the following:
- Use questions as headings to increase the chances of your content being selected as a featured snippet.
- Write content that is easy to understand and to the point, avoiding complex language.
- Use bullet points and numbered lists in your content.
- Optimize content length for featured snippets, which typically contain 40-50 words (around 300 characters).
- Add schema markup to your content to help Google understand the context and structure.
- Use synonyms of your main keywords to add context to your content and make it more likely to rank for various related phrases.
- Include images and other multimedia elements to break up the content and make it more engaging. These may also be pulling into the SERPs.
- Utilize internal linking to help Google understand the relationships between different pages on your website.
- Make sure your website is easily crawlable, adopting technical SEO best practices.
How Can I Try Out Bing Chat or Google’s Bard?
For months, Google Bard, was in beta and available only to invited “trusted testers” and staff. However, the search giant has since opened up Bard for more general experimentation in over 180 countries. Give it a try, here.
Meanwhile, Bing users can already engage with Bing Chat AI on the Microsoft Edge browser. To check it out, go here.
As more and more users experiment with AI assistants, it is likely that we will see further development in this area across multiple platforms.
Best Resources for Keeping up to Date on all Things AI and Search
- Dr Marie Haynes’ newsletter – An excellent resource from an eminent source on Google Search, Google’s Quality Raters’ Guidelines and AI chatbots.
- Digital Trends – For all things computing but regularly posts updates and insights on Search and AI.
- Lily Ray @ Amsive Digital – Also includes a really helpful take on the role of AI in Content
- Flow SEO – Our very own resources, including ChatGPT for Content Marketers
- Google Official Blog on AI
- Bing’s Official Blog on AI
“AI will have a significant impact on the content creation industry, particularly for blogs. To prepare for this disruption, focus on developing a range of unique and meaningful content assets. This might include: Thought leadership pieces; live offerings, such as in-person and online events; the development of more creators within the organization who can regularly share their perspectives; video content offerings; original research: multimedia content and animations.
When you can create anything with AI — the real question is: What do you choose to create?.” – Viola Eva – Flow SEO
Make sure to check out our roundtable discussion for more insight into AI and what it means for the SEO world!