I find the majority of predictions for the end of the year and crystal ball-gazing lists to be rubbish. They don’t back up their claims by using anecdotes instead of evidence to support their predictions, and, perhaps most important of all, they fail to assess their past predictions (how else can you tell which ones are reliable? current predictions?). Yes, I’m talking about the three of you: Gartner, Forrester, Inc., as well as Forbes. They don’t grade previous predictions however they receive the most adoring, breathless, credible reports when they write abouhowch AI will be taking all jobs in 2023, 2024, and 2030 (reality checking: AI was responsible for approximately 5 percent of US job losses in 2025) and how Google search is disappearing (reality check: not true is even as close as) or how influencers using virtual technology will replace human influencers (reality check: Brands’ investing in AI influencers decreased by 30% in the last year).
If you’re going to believe my forecasts, you’ll have to be better (and cough, cough, you’ll have to hold your sources of information to the same standards). In January, we’ll review these issues and determine how I performed.
AI usage growth is expected to slow down dramatically; the pressure on companies (and their employees) to implement AI won’t be as strong.
If you’ve been reading Datos’ State of Search reports, you’ll know that even though AI tool usage is increasing but the speed of increase is slowing. If there isn’t a major increase in the near future, AI tool use will likely slow with lower usage rates than social media or search engines, messaging apps, or commerce on the internet.
However, I’m confident that this won’t have any effect on the enthusiasm executives display for AI applications within their businesses. Teams that incorporate AI into internal applications will benefit from at the very least another year of investment that is not tied to ROI or metrics. People who work with AI will benefit from perception over those who are skeptical of AI (even when it is completely justifiable). There will be pressure to reduce the number of employees in non-AI-focused parts of companies, while huge budgets are dedicated to AI transformation.
Prediction: Corporate aggregate AI expenditure will grow in 2026 by more than 30% over 2025’s estimated $37B, and the rate of increase in consumers’ AI adoption decreases by 30 percent or more (that’s growth rate, not the level of use ).
Companies will massively invest in AI mentions and sentiment monitoring, and underinvest in nearly every other type of measurement for marketing.g
It’s not a problem that the methodology for tracking brands through AI response has more flaws than a fine mesh sieve. Don’t forget that the majority of companies that sell tracking are marketing snake oil for sale at astronomical prices (doubly that anyone could easily send commands to AI tools and then record the results). It is also worth noting the fact that there is than an order of magnitude more use of social, content, and search platforms, which businesses are unable to effectively monitor and influence.
Prediction: By 2026 2026, the tracking of AI is likely to (by itself, not including Google Search and Google AI Overview tracking) be a market worth more than $200 million. However, unlike tracking mentions of social media or Google rankings for search results and Google search rankings, it will yield nearly zero ROI (both because AI responses are by design statistical lotteries of answers and also because none of these companies has done the research to prove that their strategies are in line with what users experience on the ground). It is important to note that this will be difficult to quantify, in particular when you consider Adobe’s purchase of SEMRush.
The prevailing sentiments, not the facts, will continue to dominate the realm of beliefs and feelings.s
Be patient with me, because it’s a challenge to explain, but crucial to comprehend. Let me start by giving some examples:
- Your parents say they’re afraid to visit Ro; me, however, they say their home in Jacksonville, Florida, is much more secure. When you mention the statistics (showing the fact that Rome is a factor of magnitude more secure than Jacksonville, they’ll counter by citing anecdotes (“your cousin Johnny encountered an Italian woman who lost her purse from the station at Rome”)
- TikTok accounts advise younger New Yorkers to head to the Brooklyn Bridge. Brooklyn Bridge to see New Year’s fireworks. Thousands of people arrive and then find that there’s no fireworks at all (and they were not the only ones). It seems that fact-checking is less well-known than waiting for long hours for nothing to happen in the dark.
- Americans have a different view of the economy of their country than they did in the past, based on observable actual-world events and indicators (wages and unemployment rates, the performance of stocks, etc.). However, this is no longer the case. The perception of partisanship has always existed, but people of all parties instantly change their opinions when they see their preferred candidate/party in or out of the spotlight.
I’m not convinced that the people who took part in these surveys are fake, es like I don’t think TikTok users walked across the Brooklyn Bridge, Brooklyn Bridge, secretly aware they’d be scammed, and that Jacksonville relatives lie about their feelings of security. Humans have always been difficult to convince using data, and it is easy to make up stories.
The shift is that we now live in a world governed more by algorithmic feeds populated by strangers (usually a combination of influencers and randos-who-happened-to-win-the-algo-that-day) than by friends, family, journalism, or even lived experience. The messages generated by the stranger-feed might not be the sole source of information for the entire world; however, it is an overwhelming majority of our collective consciousness. Politicians use it to convert minds and hearts to their alternative version of the events. Brands make use of it to deflect attention from the actuality of their products and create marketing stories. People employ it to portray their idealized, self-assured, o identity-shaping versions of themselves rather than their true ones.
In the past few years, there was a widespread (though not all-encompassing) assumption that honesty would prevail. Influencers who were paid, political hypocrites, fake brands, face-tuning artists were seen as what they were: their credibility destroyed and their voices obliterated. Nothing like this has happened, and no sensible person in 2026 can imagine that it will occur in the near future. Americans (and numerous other nations) have decided, in a cultural sense and culturally, that we’re okay with fake, mostly unaffected by hypocrisy, and generally detached from the facts of our choices of entertainment, politics, or purchasing choices. Vibes, though? Vibes get engagement. Vibes get votes. Vibes drive sales. Even stock prices are influenced by the vibrations. The adage “the market can stay irrational longer than you can stay solvent,” could be applied to every aspect of life.
Prediction: By 2027, no credible data at scale will exist, suggesting that a return to authenticity, truth, or reality will be prevalent in Americans’ purchase or voting decisions, sentiment/perception, or online behaviors
The value of content written by writers will shift from engaging human beings to influencing algorithms.
Dan Petrovic’s latest research has shown us that although only 56% of B2B marketers (a large readership in the past, if there ever existed an audience like this) had “web readers” in 201,5, Half of them (28 percent)will read by 2025. In the end, I’m a fan of the research methodology just as any other I’ve seen in the field, as it cleverly blends self-reported data from surveys with real-time analytics of the behavior of visitors.
But what do they mean? Crawlers are used by search engines as well as AI tools. They scan every word, take note of every bold typeface, each word, grammatical error, li,nk, and pattern. Their summaries are a hit.
AI overviews, AI modes from Google as well as responses from AI instruments, AI summaries in ten thousand software products, as well as AI-powered feeds all undermine the importance of written content by transforming the direct, historical impact it had (i.e. people reading things) to text corpus-based models that take into the writing and transform it into small sentences we see and read.
Prediction: In 2026, Internet users will consume 10X as much content via AI summaries as they will on every page of longer-than-a-paragraph text content (even accounting for the added length of said pieces). Substacks, blog po, STS podcast transcripts, and articles in long-form, detailed paragraphs on product pages will not be 10% of the content we consume through AI.
P.S. When the marketing industry can accept this, a lot of others will write for machines versus humans (a task I personally hate and would not like to be a part of, howeverr, I do hope that the next generation will find enjoyment or at the very least, pride).
Zero-click marketing will make up the bulk of any online experience.
TikTok, Reddit, Instagram, Facebook, LinkedIn, YouTube, ChatGPT, even Google together, these social media platforms dominate the amount of time people spend online, and they also discriminate against (or completely block) external hyperlinks. Google is doing this with AI overviews and instant responses. ChatGPT does this by analyzing websites that it does not link to. The social media websites do this by implementing algorithms that limit the visibility of posts that contain hyperlinks, such as YouTube, Instagram, and TikTok, which makes links almost impossible to locate.
In the past few years, all of the top social platforms have converged on TikTok’s algorithmic feed technology: engaging content > followers accounts. In the end, the value of following a user has never been lower. Following an account or a subject that you follow on Threads, Reddit, YouTube, Twitter, LinkedIn, Facebook, or Instagram is just one percent more likely to view the content of that account or topic in the feed you use as your primary. Bluesky as well as Mastodon may be both outliers in this regard; however, their adoption rates are so low that they could be insignificant outside of a handful of specific communities.
This strategy of engaging users through followings > content and links that are zero-clicks to such an extent that the 10 top social media platforms account for over 80% of the internet time in every nation, on all devices. Social media is the place the place we receive news and information, as well as discover trends, products, and the latest brands. This is how we form our opinions. Search and AI use have increased in a dramatic way, and are sending ever smaller volumes of traffic.
We live in the Zero-Click World, but marketing remains in the fight for small, shrinking pieces of data that Google and other companies offer.
The prediction is that by 2027, the volume of traffic generated through all the major social, search, as well as AI platforms will decrease from their current levels even as the use of these apps and sites rises.
