The Future of Advertising in an AI‑Driven Landscape

***All opinions in this article are my own and not of Coty, or it’s subsidiaries. All information cited in the below article are sourced from published articles in the trade press and do not represent the experience or performance of campaigns. None of the data below is reflects internal information.***

Google’s long-reigning search monopoly is starting to show cracks. In the final three months of 2024, Google’s global search engine market share fell below 90% for the first time since 2015. In the U.S., Google dipped to about 87% share in December 2024, down from ~90% the month prior. While Google is still the clear leader, even slight drops are noteworthy after a decade of consistently holding 90–92% share. Competing engines have each captured slivers of this lost share (with Bing hovering around 4% globally). More interestingly, non-traditional search platforms are emerging – for example, ChatGPT’s AI search gained so much traction that analysts project it could reach about 1% of search query share by 2025. In fact, referral traffic from ChatGPT was growing nearly 44% month-over-month in late 2024 . This erosion of Google’s dominance signals that consumers are exploring new ways to find information and products, beyond the classic search results page.

A major driver of this shift is changing consumer behavior around product discovery. Rather than relying solely on traditional search engines, consumers are increasingly turning to AI-driven recommendations and social media as their starting points. A global survey of 12,000 shoppers by Capgemini found that nearly 3 in 5 consumers have already replaced traditional search engines with generative AI tools for shopping decisions. In particular, younger generations are leading the charge: roughly two-thirds of Millennials and Gen Z shoppers now use ChatGPT-style AI for product recommendations, compared to 44% of Baby Boomers. These AI-based interactions range from asking chatbots for the “best budget smartphone” to using voice assistants for purchase suggestions. In parallel, social media has risen as a discovery engine: about one-third of shoppers used social platforms to find or buy products in 2024 (up from one-quarter in 2023). Among Gen Z, this trend is even more pronounced – over two-thirds have discovered new brands via social feeds, and more than half have made purchases directly through social commerce channels. Even for holiday shopping, Gen Z is blending channels: during the 2024 season, 34% of Gen Z globally planned to find gift ideas via TikTok Shop, 25% via influencer recommendations, and 14% via generative AI search.

The takeaway for advertisers is that the once-straightforward path of capturing consumers via the established digital funnels is splintering. Discovery is happening in a more distributed, algorithm-driven way – whether through an AI chatbot’s personalized suggestion, an Instagram reel, or an Amazon recommendation carousel. Advertising strategies that overly depend on traditional search risk missing a growing segment of consumers who bypass search engines entirely in favor of AI-curated experiences.

Just as consumers are embracing AI recommendations, advertisers themselves are increasingly leaning on AI to manage campaigns. Platforms like Google’s Performance Max and Meta’s Advantage+ represent a new generation of automated, AI-optimized advertising. These campaign types allow marketers to simply provide creative assets and targeting goals, then rely on machine learning to decide who sees the ad, where, and when. The promise is powerful – broader reach and improved conversion efficiency through real-time optimization across channels. Many advertisers report lower cost-per-acquisition and time saved by letting these AI systems handle the heavy lifting of budget allocation, bidding, and audience targeting. Indeed, Meta Advantage+ can dynamically shift spend toward the best-performing audience segments or creative variants in real time, often outperforming manual targeting in finding high-intent users. The efficiency gains free up marketers to focus more on strategy and creative, rather than micromanaging campaigns.

However, this convenience comes with a trade-off in transparency and control. Advertisers must surrender many levers to the algorithm, effectively operating within a “black box.” There has been much discussion in marketing and media circles around “the blackest black boxes” of the AI-optimized ad products for how little is shared about the models internal decision-making. As one ad-fraud prevention CEO noted, black-box algorithms now dominate advertising, using vast data and complex models to automate ad placements – but they are “intentionally opaque”, often keeping marketers in the dark about what’s happening behind the scenes. Basic questions like Which search queries or sites did my ads appear on? or Which audience segment drove the most conversions? are harder to answer when using these fully automated campaigns. The lack of transparency into placement and conversion details leaves marketers unable to fully understand or manually optimize their campaigns. This “loss of control” can be challenging for brands with strict targeting needs or compliance requirements, since many decisions – from bid adjustments to placement selection – are handled autonomously by the platform .

For advertisers, this shift means rethinking their role and skills. Success in an AI-automated ad world relies less on manual tweaking and more on feeding the right data and creative into the system, then interpreting high-level results. It also demands new levels of trust in platform algorithms (or the use of third-party tools to audit and verify performance). The dominant ad platforms now effectively control the ad setup, delivery, and even the performance measurement, which tilts power away from advertisers. In response, marketers are learning to balance automation with oversight – for example, setting up incrementality tests and independent analytics to ensure the AI-driven optimizations are truly driving real business outcomes. This new landscape rewards those who can work with the AI (crafting better inputs and creatives) while still holding platforms accountable for transparent metrics and brand safety.

As LLMs like ChatGPT, Bing Chat, and other conversational assistants become mainstream, we’re on the cusp of another fundamental shift: searching without directly visiting search engines or websites. Instead of typing queries into Google, more users are asking AI chatbots to find information or recommend products. These AI “co-pilots” can synthesize answers from multiple sources, perform comparisons, and even execute tasks (like booking travel or adding items to a cart) on behalf of the user. The Capgemini study cited earlier underscores this trend – a significant portion of consumers, especially under 40, now use AI agents in lieu of search engines for product discovery . And as generative AI capabilities improve, it’s conceivable that an AI assistant could handle many routine search queries, delivering instant answers or personalized suggestions without the user ever scrolling through a traditional web page.

This emerging mode of search presents a paradox for the web ecosystem. On one hand, AI agents promise convenience and personalization. On the other, they blur the lines of what constitutes “real” user traffic. AI agents often fetch content via automated scripts or headless browsers, not through a user visibly clicking a link in a browser. From the perspective of websites and analytics tools, these interactions can look like bot traffic rather than human visits. Indeed, cybersecurity firms note that the web is evolving from being primarily browser-based to a mix of APIs, mobile apps, and AI agent traffic . This evolution is forcing companies to distinguish not just human vs. bot, but legitimate vs. illegitimate automated access . The challenge is that an AI agent fetching data for a user’s query is a legitimate use-case, but it’s still a non-human actor. Many websites, concerned about scrapers and data misuse, are responding bluntly – by blocking or throttling AI-related traffic outright. For instance, Cloudflare (which protects a huge swath of sites) found that its customers “overwhelmingly opt to block” even well-behaved AI bots that honor robots.txt . This means that bots like OpenAI’s GPTBot, Bing’s content crawler, or startup AI scrapers might simply be shown the door by default. Even more concerning, some AI agents have been caught trying to evade detection – one example is the AI search startup Perplexity, which was accused of impersonating a regular browser user to scrape content. Such tactics further erode trust and make sites more likely to blanket-block unknown automated visitors.

If AI agents are widely treated as “bots” and blocked, the free flow of information that powers them could be stifled. Users might find their AI assistants cannot access certain websites or are serving outdated info that was gathered before blocks were in place. In effect, the web could split between content accessible to AI assistants and content sealed off for human-only viewing. For advertisers and content creators, this raises urgent questions: How do you reach an audience that is delegating its browsing to AI? If an AI agent provides an answer without ever directing the user to your site, traditional web analytics and ad tracking break down – it’s a “zero-click” world, amplified. There’s also the risk that AI-driven traffic gets misclassified in ad metrics, leading to skewed performance reports (e.g. if ad impressions are served to an AI agent that’s crawling a page, not a human). On the flip side, some legitimate user activities might be flagged as bots – for example, a consumer using an AI shopping assistant that programmatically visits several product pages could be mistaken for a scraping bot, potentially getting blocked and harming that user’s experience. The rise of AI agents forces the industry to develop a more nuanced approach to traffic management, one that can accommodate AI-assisted user behavior without opening the floodgates to abuse.

For advertising, the advent of AI-driven search means brands may need to find ways to become visible within AI platforms rather than on the open web. We’re already seeing early moves in this direction – Microsoft’s Bing Chat, for example, serves ads within the chat interface and cites sources (offering a new kind of paid placement opportunity). If large numbers of users skip the search engine and ask an assistant instead, advertisers might have to collaborate directly with AI providers to ensure their products or messages are recommended by the AI. This could look like sponsored recommendations spoken by a virtual assistant, or integration of brand content into AI knowledge bases. It’s a future where your SEO might matter less for Google’s algorithm and more for a ChatGPT plugin or an AI’s product database. But navigating that future will be tricky as long as the relationship between AI agents and websites remains fraught, with many sites currently choosing to lock AIs out.

Facing these converging trends – waning search dominance, AI-driven consumer journeys, automated ad platforms, and AI agents – advertisers and marketers must adapt proactively. Here are key strategies and takeaways for industry leaders to consider:

• Rethink Traditional Search: Given Google’s slight decline and consumers shifting elsewhere, brands should diversify their marketing mix beyond search engines. This means investing in alternative platforms (like Amazon, YouTube, or TikTok search), and exploring opportunities in social commerce and influencer marketing, since a large chunk of discovery is happening via social feeds. It also means keeping an eye on emerging AI search apps, including Google’s own AI summaries– if ChatGPT or similar tools start directing significant traffic, ensure your content is optimized to be the one an AI references or links to.

• Embrace AI, but Maintain Oversight: Leveraging tools like Performance Max and Advantage+ can unlock efficiencies – if you feed them well. Supply high-quality creatives, robust product data, and clear conversion signals to train the algorithms. Then, complement the platform’s reporting with your own measurement. For example, run lift studies or use analytics to verify that the automated campaigns are truly driving incremental sales. Don’t be afraid to set guardrails: most black-box ad platforms still offer some controls (e.g. location or brand safety exclusions) – use them to avoid obvious misalignment with your brand. The key is to let AI handle tactical optimization while you focus on strategy and creative differentiation, ensuring the machine’s decisions align with your broader goals. Remember that AI optimizes for the metrics you give it; make sure you’re optimizing for true business value (like customer lifetime value or profit, not just cheap clicks).

• Adapt to the “Black Box” Ecosystem: Since transparency is limited, advertisers should develop new best practices for the era of automated campaigns. This could include shifting team skill sets – retraining campaign managers to become analysts of AI-driven outcomes and creators of better inputs, rather than granular optimizers. It also involves building trust, not blind trust: engage in dialogues with platform reps about performance drivers, participate in beta programs that offer more insight, and push industry groups to advocate for more transparency. Be prepared that AI-driven platforms may favor broader audience targeting; invest in strong brand messaging that can appeal to wide but relevant audiences, since micro-targeting is less in your control. As one marketing expert put it, it’s about “balancing automation with strategic control” – you might relinquish day-to-day tweaks, but you can double down on creative testing and brand strategy to guide the AI’s choices .

• Ensure AI-Readiness of Your Content: In a future where AI agents answer many queries, make your content and data AI-friendly. This could mean adopting structured data schemas so that AI systems can easily parse your site’s information, or offering APIs for approved AI services to access real-time data. It’s also wise to monitor how AI assistants present your brand or information. Just as companies now track search engine result page (SERP) appearances, you should start checking responses on popular AI Q&A platforms for accuracy and presence of your brand. If errors or omissions occur, engage with the AI developers – many have feedback loops to improve their models. Additionally, consider whether blocking or allowing AI bots on your own site is in your interest. Some publishers may choose to allow reputable AI crawlers (perhaps those that agree to certain terms) so as not to miss out on being included in AI-driven results, while still blocking malicious scrapers. Each business will need to weigh the trade-off between content protection and discovery via AI.

• Innovate in Advertising Channels: The rise of AI-generated content and recommendations opens new avenues for marketing. For instance, brands can experiment with generative AI for personalized creative – using AI to produce myriad ad variations that the platforms can test. Likewise, prepare for the possibility of ads delivered through AI assistants: we may see formats like conversational ads (where an AI suggests, “Would you like to hear about a sponsored offer?”) or product placement within AI-curated answers. Being an early mover in these formats could provide an edge. In all cases, maintain a focus on consumer trust – AI-driven marketing should be transparent and add value, or savvy consumers will tune it out (or worse, find it creepy).

Bottom Line: The advertising landscape is becoming more algorithmic and AI-centered at every stage – from how consumers find products, to how campaigns are optimized, to how information flows online. For corporate marketing leaders, the imperative is clear: adapt and experiment, or risk obsolescence. This means breaking down silos between search, social, and e-commerce teams (since AI and omnichannel discovery blur those lines), upskilling your organization in data and AI literacy, and staying agile with budget allocation as new opportunities (and risks) emerge. The companies that thrive will be those that harness AI as a force-multiplier – gaining efficiency and insight – while still steering the ship with human creativity, ethical judgment, and strategic vision. The future of advertising will belong to those who can best blend human and machine strengths to engage the consumer of tomorrow.