That old adage about time moving too quickly only somewhat recently began to resonate with me. Here we are, at the end of another year, with a handful of key AI happenings from the final month of 2025 that could signal how the New Year might unfold.
Is OpenAI CEO Sam Altman Preparing for the Worst?
The person in charge of the world’s largest AI company, Sam Altman, wrote on X recently that some AI models are becoming so “good at computer security” that they’re “beginning to find critical vulnerabilities,” which is both a good thing (for the good guys) and a big risk when the technology finds itself in dangerous hands.
In response to this and other observations about AI-related risks, Altman is urgently hiring a Head of Preparedness to try to get out ahead of any potential problems. The successful candidate will be responsible for identifying and mitigating cybersecurity threats, biological risks and psychological impacts brought on by AI. According to Altman’s post, it’s a high-stress job that requires immediate attention.
This news sparked a discussion on Reddit in a viral technology thread, with users quick to comment about their experiences, annoyances and concerns with AI models as of late.
The whole of this news hints at a 2026 where both everyday users and businesses alike begin to think more critically about how AI ultimately impacts their lives and work. Like Altman hunting for a Head of Preparedness, marketers should also be thinking about how they’re using AI within their organizations, and set themselves up with effective guidelines and guardrails that serve to protect not just their interests, but their data, employees and customers in the quarters ahead.
Subscribe to the ai marketer
Weekly updates on all things ai in marketing.
Google Rolled Out Gemini 3 Flash Globally
December 17th, 2025, marked the start of a global Search rollout for Google’s new Gemini 3 Flash, which is now available in the browser’s AI Mode. In their announcement post, Google says Gemini 3 Flash is faster, has better reasoning and is an “all-around more powerful tool” that’s capable of handling a greater degree of nuance.
Users should now be able to ask more complex questions in AI Mode and receive more comprehensive, precise answers than possible with the previous version.
As part of this rollout, Gemini 3 Pro and Nano Banana Pro are now available in Google Search as well. Just click “Thinking with 3 Pro” in AI Mode, and it will produce in-depth guidance and advanced, interactive visuals tailored directly to your query.
New York Makes Disclaimers Mandatory for Advertisers Using AI Avatars
AI-generated ads are on the rise, and more specifically, ads that feature AI-generated or altered avatars or performers. This brand-new NY bill is the first of its kind in America that aims to address just that.
Gov. Kathy Hochul signed bill S.8420-A/A.8887-B on December 9th, 2025, amending New York’s General Business Law to require advertisers to disclose when they use “synthetic performers” in their material. Here, “synthetic performer” is defined as a digitally created asset (via AI or algorithm) that gives the impression of a human performer but is not recognizable as a real individual.
There are a few important exemptions, though, including:
- Expressive works (e.g., movies, TV programs, video games) aren’t covered if the synthetic performer’s use is consistent with the work.
- Audio-only ads may be exempt.
- Situations where AI is used only for language translation of a human performer are exempt.
Failure to comply with the new law will lead to penalties:
- $1,000 civil penalty for the first violation.
- $5,000 civil penalty for subsequent violation.
For me, this is a win for ethics and transparency, and I’m curious to see how other states or nations might follow New York’s lead in 2026.
Will 2026 be the Year of Anti-AI Marketing?
Allison Morrow, a senior writer for CNN Business, published an article on December 16th, 2025, titled Why AI could be the year of anti-AI marketing, and I think it’s a fantastic take on the current state of AI, the economy and people’s waning tolerance for the technology.
Morrow writes about how it feels “crummy” to be tricked by AI videos on social media (an increasingly common thing) while highlighting the many ways in which creators and even some production companies are emphasizing the humanity behind their works, while other big brands brook harsh public criticism for sloppily adopting AI or offering weak, buggy, or incompetent AI features or products.
Right now, it seems 2026 will be almost like a decisive game of tug-of-war between the pro-AI folks versus those who are staunchly against it to determine how the world will ultimately move forward in the years and decades ahead. Will the AI bubble burst? Will people simply begin to ignore it and carry on as if it were pre-generative AI 2022 again? I think 2026 will be an important year for determining how we move forward.
Google’s AI Co-Scientists Show Promise
To end with a lighter entry, Google is continuing its work to get advanced AI models in the hands of American scientists to accelerate discoveries across biomedicine, physics, chemistry and computer science.
In December, Google began providing an “accelerated access program for scientists” at all 17 Department of Energy (DoE) National Laboratories across the U.S. They say it’s already shown some promise, and that they plan to expand the program in 2026 with more tailored AI tools for material science, drug discovery, energy, genome biology and even a family of weather forecasting models.
Amid all of the uncertainty and wildly mixed opinions surrounding AI, use cases like these that help humans make breakthrough scientific discoveries quicker are among some of the tech’s objectively positive benefits. Ideally, Google’s growing work here with the DoE will contribute meaningfully to many around the world.
Final Thoughts
At present, tension seems high between those who love and loathe AI. Among the latter, tolerance also seems to be shrinking. I think 2026 will be an important year in terms of how we ultimately decide to move forward with AI, but the volatility we’ve become used to over the past 365 days makes it difficult to predict the future with a high degree of certainty.
