The State of Generative AI Strategy in 2024

A year of new model releases raised the bar

2024 opened with a wave of new AI model releases that materially changed what businesses could realistically do with generative AI. Google released Gemini 1.5 in February 2024 with a dramatically expanded context window, letting it process much larger documents at once. Anthropic released the Claude 3 family, Opus, Sonnet and Haiku, in March 2024, and OpenAI followed with GPT-4o in May 2024, adding faster, more natural multimodal capabilities. For businesses, this meant the tools available in December 2024 were meaningfully more capable than the ones available in January, which made generative AI strategy a moving target rather than a decision made once and left alone.

From novelty to workflow

By 2024, most marketing teams had moved past the early experimentation phase, poking at ChatGPT out of curiosity, and into figuring out where generative AI belonged in a real workflow. OpenAI’s custom GPTs, launched in late 2023, matured through 2024 into a common way for teams to build narrow, purpose-specific assistants, a brand voice checker, a customer FAQ drafter, without needing engineering resources. Surveys of marketers throughout the year consistently showed a majority using generative AI tools in some part of their work, though far fewer had a formal policy governing how it should and shouldn’t be used.

Disclosure and trust became real strategic questions

As AI-generated content and AI-assisted advertising claims became more common, regulators and platforms started paying closer attention. The US Federal Trade Commission continued sharpening its guidance through 2024 on disclosure requirements for AI-generated endorsements and reviews, warning that deceptive AI-generated testimonials could trigger enforcement action. For businesses, this turned “should we disclose AI involvement” from an ethical nice-to-have into a genuine compliance question worth building into any generative AI strategy rather than treating as an afterthought.

Google drew a hard line on low-quality AI content

Google’s March 2024 core update specifically targeted what it called scaled content abuse, sites using AI to mass-produce low-value pages purely to rank. Businesses that had rushed into AI content generation as a shortcut to publishing volume saw real ranking damage, while businesses using AI as a drafting aid inside a genuine editorial process were largely unaffected. That distinction, AI as an accelerant for real expertise versus AI as a replacement for it, became one of the defining lessons of the year.

What a coherent strategy looked like by year’s end

The businesses that ended 2024 with something resembling a real generative AI strategy, rather than scattered, ad hoc tool use, tended to share a few habits: a clear list of approved use cases rather than blanket permission, a human review step built into any customer-facing output and a way of tracking whether AI use was saving time or improving results rather than just feeling modern. Businesses without that structure spent much of the year cleaning up avoidable mistakes, from off-brand copy to content that quietly damaged their search visibility.

Looking toward 2025

By the end of 2024, generative AI had stopped being a side experiment for most marketing teams and started becoming genuine infrastructure, embedded in drafting, research and even customer service workflows. The strategic question shifted from whether to use these tools to how to govern their use responsibly, a shift that set the stage for the more mature, more scrutinized adoption businesses would need heading into 2025.

---

More from 2024: additional AI marketing guides, the rest of 2024 on this site or the Digital Business Marketing Awards.