In a not too distant future, it will be common practice for everyone to have their own personalized AI agent to field emails, share news or plan vacations.

Why it matters: If artificial intelligence-supported chatbots begin to dominate people's attention, communicators and marketers will seek to influence the chatbots' conversations.

The big picture: This could be the next mutation of search engine optimization (SEO).

• Because consumers, employees and investors are increasingly relying on large language models (LLMs) — like ChatGPT or Google's Gemini — to answer questions or give recommendations, brands are increasingly concerned with how these tools are representing them.
• Companies want to make sure their brand-approved language is used and that they are not associated with problematic messaging or narratives.
• For example, financial firms do not want LLMs to associate them with terminology like "ESG" or "woke" during an election year, per Fast Company.

What they're saying: When fielding inquiries about businesses, events or historical figures, LLM's responses come with higher stakes given the potential for bias or misinformation, digital marketing strategist Chris Gee told Axios.

• "If you ask AI to tell [you] about the labor situation at Starbucks, what is it going to say? That answer is going to influence how people might think about that brand's reputation moving forward," Gee said.

Some brand marketing specialists are looking for ways to influence LLMs with accurate — and positive— information about their company or product, according to Gee.

• "[They] need to be able to define exactly the type of information that goes in, because if someone asks ChatGPT for a restaurant recommendation, for example, and [their]restaurant is withholding its information and data, then it's not going to recommend it."

Zoom in: Reddit is experiencing a similar phenomenon, according to a 404 Media report.

• After Google announced it would use Reddit content to train its LLMs, AI-powered services — or bots — started popping up on the platform to automatically mention and plug products across appropriate Reddit threads.

What to watch: There's ongoing discussion in the marketing and communications space around the ethics of manipulating or influencing AI outputs.

• "The wheels are up and it's inevitable that AI becomes as mainstream as SEO," says brand and marketing expert Jen Kling. "But with that could also come a massive stream of data chaos, lack of oversight, inherent bias and the potential for more mis-disinformation."

💭Thought bubble from Axios AI+ author Ryan Heath:

• "AI is transformative, but the emergence of an industry dedicated to shaping AI models to reflect corporate PR positioning is not an obvious transformation Silicon Valley engineers plan for. This is another reminder of why broad conversations are needed about AI rules and guardrails."

Share this story

The more I reported on this concept, the more complex it felt, so naturally I turned to the source for answers.

Prompt: Tell me how a brand can influence how it is perceived by ChatGPT.

Response: "A brand can influence how it is perceived by ChatGPT through various means, including:

• Consistent messaging: Maintaining a consistent brand voice and messaging across different channels can help ChatGPT develop a clear understanding of the brand's identity and values.
• Quality content: Creating high-quality content that reflects the brand's expertise, authority and trustworthiness can positively influence how ChatGPT perceives the brand. This includes informative articles, engaging social media posts, and well-designed websites.
• Positive interactions: Engaging with ChatGPT in a positive and helpful manner can foster a favorable perception of the brand. This includes providing useful information, answering questions, and addressing concerns in a timely and respectful manner.
• (Note from Eleanor: This response threw me for a loop!)
• Brand associations: Building positive associations with other reputable brands or influencers can indirectly influence how ChatGPT perceives the brand.
• User feedback: Listening to user feedback and incorporating it into the brand's messaging and strategy can help improve how ChatGPT perceives the brand. This includes addressing customer concerns, implementing suggestions for improvement, and demonstrating a commitment to customer satisfaction.

Overall, by focusing on consistent messaging, quality content, positive interactions, brand associations, and user feedback, a brand can effectively influence how it is perceived by ChatGPT and, by extension, its audience."

Communicators are more likely to have a favorable view of agencies and colleagues who incorporate AI into their work, according to an Axios-Seven Letter insights survey.

Why it matters: This signals that AI-supported work might not be as taboo as it was just one year ago.

By the numbers: Seven Letter surveyed Axios Communicators readers against general working professionals and found that communicators are more likely (87%) than the average professional (62%) to use AI in their everyday work, specifically when it comes to generating new content.

• Overall, most professionals are using AI for research support, per the survey.

Zoom in: Communication pros are most concerned about the accuracy of AI, and all of those polled say they edit AI-generated content before shipping it off.

• Communicators are also more likely (86%) than average professionals (73%) to want to know how professional services firm are using generative AI to enhance their services or products.

What they're saying: "Professionals are using generative AI cautiously and see the potential it can deliver to society, to their own work and to the success of their firms," says Matt George, partner and head of research at Seven Letter.

• "At the same time, professionals want clear communication about how the firms they interact with are incorporating generative AI into their work product."

Share this story

As part of the AI hype wave, many communications and advisory firms are rolling out their own proprietary AI tools to peddle to clients.

Why it matters: The effectiveness of the tools and the ability for firms to quickly upskill its workforce remains to be seen.

State of play: Advisory firms like BCW and Teneo are using AI to monitor global trends and for predictive measures, like treating AI as a proxy for a very specific audience and then message testing against it.

• Edelman has woven its Trust Barometer data into a proprietary AI tool called Archie, which tracks how trust is moving and provides real-time recommendations to improve trust.
• Gregory FCA was quick to roll out a free AI press release generator in early 2023 and since then has created personalized GPTs for media monitoring, media training and finessing or editing content.

Public relations agency SixEastern is getting super tactical by partnering with its client Perplexity.ai to embed the "answer-engine" into its PR offerings.

• Firm founder Emilie Gerber says it will soon use Perplexity for news monitoring through "daily news reports" — which could change the way PR firms engage with costly media monitoring platforms and research tools.

What to watch: How these AI-supported PR efforts are communicated internally and to clients is a work in progress.

• According to a Seven Letter insight survey, only half of communication professionals report having internal policies and reporting requirements for AI use.
• Meanwhile, roughly 9 in 10 communicators want clear records of how firms are using generative AI in their client work.

Read more

❓How is your firm using and reporting its AI use? Shoot me a note.

AI is forcing corporate boards to change how they operate, with the most aggressive companies appointing AI bots as observers to their boards and putting tech at the center of their board strategy work, writes Axios' Ryan Heath.

Why it matters: New uses of AI could support boards to make more informed decisions.

Yes, but: Most boards lack the expertise to effectively guide AI strategies.

• The pace of AI innovation is pushing executives to develop AI strategies quicker than they have for previous new technologies, and boards are now part of that wave of change.

By the numbers: Only 13% of S&P 500 companies have at least one director with AI-related expertise, according to an analysis of SEC filings by ISS-Corporate, a governance consultancy.

• Just eight companies in the S&P 500 have formal AI board oversight systems, per ISS, and only around 1 in 7 Fortune 500 boards have a broader science or technology committee.

Keep reading

As global head of communications for Canva, Fenot Tekle is responsible for leading the graphic design platform's corporate narrative and storytelling efforts across the globe.

• Why it matters: AI-supported tools — like Canva — have expedited the shift from text heavy communications to visual communications, like graphics, data charts, photos and videos.

🗣What she's saying: "The world is becoming very, very visual and it has definitely changed the way we all communicate with each other," Tekle told Axios.

• "So much of what we're doing is through the screen. ... To be able to actually communicate visually is a value add in terms of team collaboration, productivity and efficiency, and we're really moving away from a world of text in many ways. So, for Canva, we see an opportunity to really drive that trend, because the future is poised to be more and more visual."

📍How she got here: Tekle was part of Netflix's early communications team and helped establish the streamer's brand. She then held various communications roles at LinkedIn, Facebook and Adobe before joining Canva in 2022.

🏗 How's it structured: She reports to Canva chief marketing officer Zach Kitschke and oversees a global team of 16 responsible for corporate, enterprise and product communications, analyst relations, issues management, talent brand and media events.

📈 Trend spot: Customer advocacy replacing "influencers."

• "We've seen a ton of influence come from the advocacy of our most enthusiastic users, and we want to bring them into the fold," says Tekle.
• "Companies are quick to write big checks for big-name influencers to publish a couple of social media posts, but there's really nothing like active users who've really embraced the platform, and if you can get enough of them to authentically spread the word about your product, then you're golden."

🧠 Best advice: Don't spend too much time looking in the rearview mirror.

Go deeper ... Read the entire spotlight

Good news! Journalists are replying to pitches at record-high rates, according to the latest report from public relations management platform Propel.

• Bad news: That rate is just 3.4%.

Zoom in: Tuesday is the most popular day to pitch, while most journalists reply to pitches sent on Thursdays.

• Fewer pitches were sent in the first quarter of 2024, which suggests PR pros are relying less on "spray and pray" tactics, per Propel.