Every six months, a sharp look at AI and Search reveals the landscape has shifted yet again. The pace of change is so relentless that a monthly review might soon be necessary. The first half of 2026 saw money, traffic, jobs, and market cap move before anyone could prove how much value AI actually created. Search behavior changed, token budgets exploded, software stocks sold off, and companies routinely blamed layoffs on AI. The common thread running through every major story in H1 2026 is attribution: we are still trying to measure what is happening, why, and what it means.
Why Measuring AI Visibility Has Become a Fragmented Nightmare
One of the hardest questions marketers face is how to track a brand’s presence in AI Search. The answer is that it is highly complex. You must factor in the specific engine, personalization settings, reasoning levels, model updates, and stochastic variability. The measurement landscape is fragmented. A key finding from recent user behavior studies is that the overlap of citations between engines is minuscule: 91% of citations appear in only one of ChatGPT, Perplexity, or AI Overviews. Prompt tracking should therefore be understood as closer to polling and focus groups than traditional SEO rank tracking. You cannot treat it like a keyword list.
Brand mentions in AI answers are proving more impactful on business outcomes than simple citations. Citations shape the answer, and for some businesses that is most important, but the majority of vendors and merchants need to pay attention to how often they show up in a panel of prompts, in what context, with what sentiment, and whether they are recommended ahead of competitors.
How Trust Dictates User Choice in AI Recommendations
Trust is the highest currency in AI search. Close to 75% of consumers pick the number one result in an AI shortlist. However, if they see a trusted brand anywhere on that list, they will pick it instead. The average U.S. adult already has high trust in AI recommendations. In one study, 88% of users accepted AI Mode product recommendations as the best available. In AI Overviews, by contrast, users click, evaluate, and compare a lot — that is classic search behavior. It does not carry over to AI chatbots, where users are far more willing to accept the first suggestion.
What does this mean for optimization? Writing style, token budget, and technical factors matter for agents. If you want to be present in AI Search and agentic recommendations, you must lead with unique information, write in a direct, easy-to-understand way, remove fluff, and keep your site technically fast and easy to access. AEO and GEO are brand channels disguised as performance channels. AI recommendations shape demand, not citations. The unit of optimization is whether AI names, trusts, and recommends your brand.
The Token Explosion: When CEO Mandates Turned Into Workslop
November 2025 marked a pivotal turning point. Claude’s Opus 4.5 was the first model perceived as reliable and robust enough for agentic workflows. Shortly after, Peter Steinberger’s Clawdbot went viral, leading to millions of installs, long lines in China, and even Nvidia building a Nemoclaw clone. Then things got even crazier. Shopify, Uber, Meta, and many other tech giants built token leadership boards that incentivized engineers to burn as many tokens as possible.
Jensen Huang captured the spirit of the moment with a thought experiment: “Let’s say you have a software engineer or AI researcher, and you pay them $500,000 a year. At the end of the year, I’m going to ask him how much did you spend in tokens. And if that person said $5,000, I will go ape something else. If that $500,000 engineer did not consume at least $250,000 worth of tokens, I am going to be deeply alarmed.”
A Meta engineer created an internal leaderboard dubbed “Claudeonomics” that ranked over 85,000 employees by the number of AI tokens they burned. It led to massive waste as employees left AI agents running on idle or useless tasks just to climb the ranks and earn titles like “Token Legend.” Reportedly, Meta engineers consumed 73.7 trillion tokens in just 30 days. It is hard not to wonder how much of OpenAI’s and Anthropic’s growth was ignited purely by workslop.
Good times. In April, CFOs pumped the brakes after discovering that employees had burned through their annual token budget in four months. TokenMAXXXING shifted to valueMAXXXING. Meta’s leaderboards were shut down in April, along with similar boards at other companies. But not before the number one user on the board averaged 281 billion tokens, which would cost over $1.4 million at standard API rates.
The Real Productivity Gains That Survived the Token Tsunami
What remained from the all-you-can-eat token craze is a step change in productivity that we cannot yet fully measure. AI changed work patterns fundamentally. People used AI to raise their impact by a magnitude: designing landing pages end-to-end that made it into production, automating topic prioritization, SEO testing, and SEO reporting with full-blown apps, and building arrays of useful applications from automating internal reports to building agents for research and data visualization. One outcome of the token tsunami is that a lot more people started using Claude, which in return fired up the growth engines for AI infrastructure companies that Claude recommended, like Supabase.
The SaaS Bloodbath: Market Cap Wiped Out by Perception of AI Risk
In February 2026, following the Opus 4.6 release and Claude Cowork, the software-as-a-service industry experienced a severe market downturn that wiped out hundreds of billions of dollars in market cap over just a few trading sessions. The SaaS vertical saw a steep annual decline of more than 30% in the markets. Valuations for high-growth software companies collapsed, with revenue multiples shrinking dramatically compared to their pandemic highs.
If you work in B2B, you have likely felt the impact: lower conversion rates, longer sales cycles, and softening brand searches. It turns out not all software was forsaken. The decline is purely associated with how the market views a company’s AI robustness. Company valuation drops are not related to performance, but rather to whether people think AI will disrupt their product. Over the last 30 trading days, both the top quartile and median IGV stock have outperformed the ETF as a whole. It is the bottom quartile, which includes some of the largest companies, that pulls down overall performance. If you are in B2B and looking for a new job, pick a company the market is optimistic about.
AI Washing in the Labor Market: Layoffs Blamed on a Cause That Has Not Arrived
You have likely read the headlines: “We’re laying off x% of our workforce due to AI.” It is the narrative du jour. Challenger, Gray & Christmas reported that AI was the leading cited reason for May 2026 job cuts and had been cited in 87,714 cuts year-to-date, equal to 22% of all 2026 announced layoffs through May. The numbers are real: tech layoffs up roughly 66% year over year toward 150,000, Oracle cutting 21,000 citing AI, Block cutting 40% and seeing its stock rise.
The only problem is that AI is not the cause. A lot of these layoffs were really to offset capital expenditures, pandemic overhiring, and general economic turbulence. The Big Labs have used AI replacement as a marketing narrative, but it has not come true yet. We can all imagine a future in which AI is so good that it makes certain roles obsolete, but that time is not now. Despite clear productivity gains from AI, the recent waves of layoffs show no indication of AI being the underlying cause. In fact, companies using AI as a reason to lay employees off are rehiring them.
The Agent Market Fragments: ChatGPT’s Dominance Slips as Google Rises
ChatGPT’s market share decreased from 78% in July 2025 to 56% in July 2026, while Gemini went from 15% to 30% and Claude from 2% to 10%. Google is now the most likely player to win the AI consumer market. ChatGPT still has roughly 1.1 billion users, but OpenAI has refocused on an enterprise pivot, shutting down Sora, dropping video in ChatGPT, and killing Instant Checkout. Enterprise makes up about 40% of revenue, heading toward 50% ahead of a potential IPO. According to Ed Zitron, OpenAI spends $2.8 billion every month to make $1.1 billion.
Anthropic’s fight with the Pentagon pushed Claude to number one among free apps on Apple’s U.S. App Store. A few months later, users seem less sure if Anthropic can resist attacks from open-source models. The Big Labs massively subsidize tokens. SemiAnalysis found that a $200 plan for frontier models gives you between $8,000 and $14,000 worth in tokens. Lately, open-source models like Kimi K3, GLM 5.2, and Deepseek V4 have put pressure on the big US labs because of those subsidies. Open-weight and open-source models also show that the model itself becomes commoditized, while the harness and application layer gain more importance.
Forward-deployed engineer job postings rose 800% in the first nine months of 2025 and were still up 729% year over year in April 2026. Frontier Labs now pays more than $500,000 in total compensation for people who can turn a model demo into a working system inside a customer’s business. Never have marketers faced a faster evolving environment with more platforms to measure. Model availability has become a business risk, which is why understanding which platforms matter most for your audience is critical.
Publishers and the Law: The Content-for-Traffic Model Breaks Down
Earlier predictions that up to 70% of 2024 organic traffic could be gone by 2026 were overeager. The actual figure lands at roughly 33% referral loss year over year, with 68% of Google searches now ending without a click. Publishers are moving the fight to the courts and regulators. USA Today CEO Mike Reed explained that they are getting close to the point where they will block Google and abandon traditional search traffic entirely.
Lawsuits and regulatory actions have accelerated. The U.S. Copyright Office is reviewing fair use in the context of AI training data. Major publishers including The New York Times and The Intercept have filed suits. In response to the intense publishing landscape changes, Parallel AI and Cloudflare are both trying to build publishing marketplaces for content owners and AI companies. Conde Nast CEO Roger Lynch summed up the new reality succinctly: “Assume there’s no search. You have to have your businesses planned as if search is zero.”
Publishers are trying to replace the content-for-traffic bargain with a content-for-training market. This leaves several open questions: Who sets the price and how? How can you prove that an article contributed to an AI agent’s answer or a purchase? Do publishers have negotiation power? Are smaller publishers even needed by AI labs? The answers are not yet clear, but the direction is unmistakable. The old bargain is broken, and a new one has not yet been settled.
What remains certain is that the economic impact of AI is expanding faster than our ability to attribute it. Whether you are a marketer trying to measure brand mentions, a CFO trying to justify token spend, a publisher trying to protect content value, or an investor trying to gauge market risk, the challenge is the same: the ground is shifting under our feet faster than we can map it. The companies and individuals that thrive will be those that build measurement frameworks as adaptable as the technology they seek to measure.
