Can we make AI advertising privacy safe?

Or going beyond the press releases

It’s no longer up for debate that AI is going to reshape the advertising operates. But unless we’re going to have another round of Cambridge Analytics style scandals, we need AI to not only be privacy safe, but to take a privacy first approach.

In just the last week, announcements from IBM and Anthropic show that the conversation around AI is shifting toward privacy and safety—but are these moves really about consumer protection, or are they playing to each company’s strengths? As more organizations adopt AI-driven advertising models, the question becomes: is the industry genuinely addressing privacy, or is it a convenient marketing strategy for companies keen to emphasize their specific advantages?

Most ad tech players are rushing to implement hyper-personalized targeting and fully automated workflows. But with these capabilities comes a heightened risk of data misuse and privacy violations. IBM and Anthropic’s recent announcements suggest they’re leading the charge in making AI safer. However, a closer look reveals these steps might be less about a genuine concern for privacy and more about bolstering their competitive positioning. The question remains: will the rest of the industry follow suit—or just pay lip service to the growing privacy concerns?

The Hidden Risks of AI-Driven Advertising

AI models in advertising are becoming increasingly powerful, but with great power comes great responsibility—or at least, it should. Today’s AI systems, used to optimize targeting, automate campaign management, and offer data-driven insights, require access to massive amounts of user data. This presents significant risks around how that data is collected, processed, and safeguarded.

AI relies heavily on real-time data streams, which makes it efficient—but also vulnerable. From audience segmentation to performance tracking, advertisers have to collect and process vast amounts of personal data. With this level of dependence on user data, privacy violations become more than just theoretical—they’re a daily risk. And while AI is making advertising smarter and faster, it’s also raising serious questions about whether the industry is prepared to protect consumers.

IBM and Anthropic: Playing to Their Strengths

Let’s take a closer look at the IBM and Anthropic announcements. IBM has made a play with its Granite models, which tout “enterprise-level security” as a key feature. Is this a groundbreaking move toward safeguarding data, or just IBM leaning into its existing reputation for enterprise solutions? For a company like IBM, which is already deeply embedded in enterprise security, talking about privacy in AI is less a bold step and more a logical extension of its brand. After all, IBM thrives on being the go-to for companies worried about data security. So of course, they’ll position their AI offerings as privacy-first. It’s smart business, but is it genuinely transformative?

Even through Anthropic’s Responsible Scaling Policy introduces thresholds for AI safety—especially around preventing autonomous AI from going rogue, Anthropic’s entire business model is built around “AI safety,” so it’s no surprise that they’re positioning themselves as the ethical choice in AI development. But their approach raises a question: is this focus on safety driven by a sincere desire to mitigate risks, or is it a calculated effort to carve out a niche in an industry where most players are racing toward more powerful, less regulated AI? Their framing as the “responsible” AI company certainly makes them stand out—but it’s also conveniently aligned with their core identity.

The Balancing Act: Innovation vs. Responsibility

For the broader advertising industry, the challenge is clear: how can we adopt the full potential of AI without sacrificing user privacy? It’s not enough to simply implement AI that makes processes more efficient or targeting more accurate. As AI integrates more deeply into the advertising ecosystem, the risks surrounding data privacy and security will only intensify.

The real question is whether the industry will take privacy concerns seriously, or continue to frame privacy as a secondary consideration. Companies like IBM and Anthropic are showing that AI providers can—and should—prioritize privacy, but their moves highlight an industry-wide trend: protecting user data is now a selling point, not just a legal obligation.

However, focusing on privacy isn’t just good PR; it’s a necessity. Consumers are more aware than ever of how their data is being used, and platforms that fail to take privacy seriously will face increasing scrutiny. But as long as privacy remains a buzzword rather than a guiding principle, the risk remains that companies will only do the bare minimum to meet regulatory standards, rather than taking meaningful action to safeguard user trust.

Conclusion

The real test will be whether other key players—like OpenAI Google, Meta, and Amazon—start taking similar steps, not just to improve their public image but to genuinely protect users. These companies play an even larger role in AI-driven advertising, and if they don’t follow through with concrete actions, the industry as a whole will continue to face serious privacy challenges.

The future of advertising is AI-driven, but it needs to be privacy-first. Not because it’s trendy, but because it’s critical for the survival of trust in digital advertising. The question is whether the rest of the industry is ready to embrace the privacy revolution, or whether these discussions will remain more about marketing than meaningful change.