New tools and techniques for PR and corporate comms teams to harness the power of AI
We’re helping PR and communications teams move fast to adopt the new strategies, content skills and technologies they need in a world where AI is rapidly changing the way people discover and search for information.
Shooglebox AI is designed for some of the biggest current challenges and opportunities for PR professionals navigating the rise of AI, the evolution of social media into interest-based media like TikTok and Reddit, and the decline in influence of mainstream news media.
The role of PR and corporate comms in the age of AI
AI is already transforming the way people get their information and form opinions. AI systems like ChatGPT and Google’s Gemini have become intermediaries, deciding what's surfaced, how it's framed and what gets left out. The dominant role of search engine results with links to news articles and other sources is increasingly being taken over by AI systems that gather, summarise and present information directly to the user with little need to click through to sources.
The declining traffic from search engines is having a big impact on already struggling news media, who themselves have turned to AI for solutions as they continue to cut the number of journalists they employ.
It’s the job of PR and corporate communications teams to understand and to influence how and what the AI models are surfacing – and how to use AI to arm themselves with the tools and techniques to do it.
Influencing AI answers is “a business-critical reputational function”
As public relations consultant Stuart Bruce – aka the PR Futurist – wrote recently: “AI tools are now de‑facto intermediaries. They mediate reputation in the same way journalists once did … If AI‑generated answers become the primary way stakeholders form impressions, then influencing those answers becomes a business‑critical reputational function.”
​
Bruce was reacting to the Gartner report Top Predictions to Inform 2026 Comms Strategies which forecast that by 2027 the mass adoption of LLMs as a replacement for traditional search will increase the importance of the role of communications specialists and require radically different tools and techniques from those used in traditional SEO.
​​The report said: “While traditional SEO efforts are best served by marketing, the growing demand for answer engine optimization (AEO) to build visibility and reputation requires communications-specific skills to balance stakeholder trust and platform requirements.”
AEO, GEO, narrative intelligence – solutions or snake oil?
PR and corporate communications leaders are well aware there is no shortage of snake oil salesmen seeking to quickly reposition their SEO offerings with new buzzwords and acronyms like GEO (generative engine optimisation) and AEO (answer engine optimisation).
​
Communications and marketing directors’ inboxes are bombarded by emails from SEO companies claiming they’ve got the answers to influencing the LLMs in the same way they’ve spent years using tried-and-tested ways to influence Google search results.
If it’s not that, it’s software companies rapidly trying to dress up their social listening tools as AI-powered “narrative intelligence” platforms. Brands’ PR and insights directors learned long ago that the effectiveness and value of most social monitoring platforms fell far short of their expensive promises. Marketing teams were seduced by salesmen showing them the fancy graphs, charts and statistics they could have at their fingertips – while PR and insights teams looked at the actual results they delivered and saw they had limited practical use. The vast majority of content and conversation across Facebook, TikTok, Instagram and other social platforms was inaccessible to the social listening tools – and the superficially useful-looking graphs masked the fact that a lot of what passed for “social sentiment” was based on posts on X (formerly Twitter), abandoned long ago by most people brands want to reach and understand.
Now those same salesmen are touting their platforms as all about narrative intelligence – describing it with phrases like “the ability to uncover and understand the stories” behind the quantitative data. Even the Gartner report mentioned above talks about 45% of CCOs predicted to adopt narrative intelligence platforms by 2029. But what does narrative intelligence look like? The fancy dashboards the salesmen are currently showing as the answer to “tapping into cultural trends” often look suspiciously similar to the bubble charts and phrase clouds delivered by their old platforms they now dismiss as “traditional” social listening and media monitoring.
And perhaps unsurprisingly they position their narrative intelligence platforms as being even more useful to brands’ marketing strategists and consumer insights teams – the people with much bigger budgets than the PR teams whose ability to manage brand reputation and perception relies on knowing what their target audiences are really seeing, thinking and saying. The communications directors we speak to aren’t falling for it.
​
So what’s different about Shooglebox AI?
Shooglebox AI is designed to help PR teams in a number of priority areas
Shooglebox co-founder Tony Cuthbertson says: “We’re working closely with some of the UK’s biggest brands to rapidly deploy highly practical AI solutions for PR teams that combine technology knowhow with hands-on editorial and reputation management experience.
“Shooglebox AI combines the rapidly evolving capabilities of the leading LLMs like Google’s Gemini, OpenAI’s ChatGPT, Anthropics’ Claude and X’s Grok with our own constantly evolving AI features purpose-built for some of the specific opportunities and priorities PR leaders know they need to move fast on now.
“It’s not about passing control to machines or expecting LLMs to replace the skills, experience and judgment of hardened communications professionals. It’s about radically speeding up tasks that can take humans so long they’re beyond the resources and capability of any one team. And about quickly uncovering patterns, themes and surprises among huge amounts of content and conversation. And it’s using the ability of AI to quickly apply best practice in structuring and attempting to summarise that – so human skill can be used for the vital job of truly interpreting and understanding those valuable hidden narratives and deciding how to act on them and what to say about them.
​
"That's what Shooglebox AI is designed for – it's like a superpowered assistant for your PR team capable of analysing, spotting themes and drafting summaries from large amounts of social posts and comments the native LLMs struggle to access unless you give them the posts – and create draft outputs in a style and structure that's easy to read and understand both by people and by AI systems."
It may sound obvious … but AI answers are only as good as the content each LLM can find. We’re helping PR teams identify the many gaps in content AI encounters when people are asking different kinds of questions that can shape perceptions of their brands, products and people.
The aim is to identify and address:
Disinformation:
Content designed to mislead – often in a coordinated way that can lead AI to take high repetition as proof
Misinformation:
Unintentional mistakes and misunderstandings amplified and recycled in AI answers
Missing information:
Content AI can’t see or understand – including the vast majority of posts and comments on platforms like TikTok, Facebook and Instagram
In the age of AI a company’s owned media, including corporate newsroom and information websites, takes on a renewed strategic importance as a source of useful and relevant content. But only if the style and structure of the content is designed with the needs of the LLMs in mind, as well as the ultimate audience.
Some of the latest new features in Shooglebox AI are designed to rapidly speed up the process of creating large amounts of high quality discoverable content written in a style and structure that makes it easy for AI to interpret, understand and integrate into answers to potential questions.
In just a few steps you:
1
Pull your source content into a stack in Shooglebox
2
Ask AI to analyse it and add to it or enhance it with more content from external sources
3
Tell the Shooglebox AI Assistant what kind of output you want – using a combination of pre-canned prompt engineering and your own guidance
The pre-canned prompts are very detailed sets of instructions designed, stress-tested and constantly updated by our own team of highly experienced writers, editors and software specialists – and we work with brands’ PR teams through initial set-up and support to help them customise the prompts to their own brands’ needs.
In the case of AI discoverable content the pre-canned prompts are designed to help create articles that are clearly structured and written in a way an LLM will see them as highly relevant, credible and useful for extracting and incorporating into their own answers to questions people are asking.
The end result is a draft a skilled PR professional can use as the basis for a final article, complete with clickable links to online sources and structured in a way it’s ready to import into the company’s own CMS.
It’s no substitute for the judgment, writing flair or audience awareness of a journalist, editor or PR professional but it takes a helluva lot of legwork out of the job – and reduces to minutes tasks that previously took hours or days.
And the same tools – with different detailed pre-canned prompts – can be used to help with all kinds of other outputs:
​​​
-
​Internal briefing notes on current trends​
​
-
Reports on reputational issues and reaction on social media​
​
-
Consumer-facing blog posts or articles​
​
-
Press releases and materials for news media
2. Gathering, analysing and summarising conversation on social media
Some of the core features of Shooglebox AI are designed for gathering, analysing and summarising posts and comments on platforms like TikTok, Reddit, YouTube, Instagram and Facebook.
One of the biggest time-saving innovations is the ability to automatically enhance card backs by capturing lots of material from social media posts including:
-
Video transcription – Word-for-word transcripts of what people are saying in their videos, with language translation if necessary
​
-
Themed comments – Pulling in comments with the option for AI to have a first pass at selecting the most useful/relevant and sorting them into themes
​
-
Image description – Analysing what pictures show and turning captions and overlays into editable text
​
-
Profile info – Details on the poster including bio, account links, follower and engagement statistics
It means brands can quickly uncover and understand the different things people are saying about their products, services or important topics of interest in a way that can be turned into actionable insights, summaries or articles with the help of the Shooglebox AI Assistant.
Example: What people are saying about latest menu changes at McDonald’s
You can see an example of just how useful this is in this blog post about reaction across social media to McDonald’s finally bringing the American breakfast favourite the McGriddle to the UK. McDonald’s regularly refresh their menus throughout the year – so there’s a constant stream of new talking points around the brand on social media as thousands of people review and tell others about the new items.
Shooglebox AI does in a couple of minutes what would take a human researcher or analyst several hours. The theming of thousands of comments on a popular post can be done in seconds – and hundreds of posts can be analysed simultaneously.
The AI models are trained using a carefully defined set of instructions that outline what high-quality, relevant comments look like across different types of conversation topics, optimising selection towards the highest-value comments. AI guardrails ensure comments are quoted verbatim, preserving wording, tone and emojis exactly as written and preventing the tendency of AI to hallucinate.
Why AI LLMs struggle to see most content on social media
What makes this invaluable is that native LLMs like ChatGPT, Gemini and Claude struggle to access the vast majority of content and conversation on social media.
It’s the same with social listening platforms. Without being able to log in and access most content on social platforms, listening tools are unable to see the most useful conversations about a topic, which are often in videos and comment threads on platforms like TikTok, not posts on X and services where social listening tools have some kind of API access.
It means AI tools often have to rely on a few news articles or blog posts about social media conversations – and these may give a deliberately skewed picture of what people are saying about a brand, a product or an individual.
Shooglebox AI means PR teams can not only quickly get to the heart of multiple conversations and talking points across social media but also create AI discoverable content that helps give AI sight of hidden conversations with the context to understand them and links to sources as validation.
3. Decoding cultural trends and tapping in to interest-based communities
The evolution of social media towards interest-based media means culture is no longer driven by mass media and the mainstream – it’s shaped by the millions of people creating content and talking about their passions and interests in different hyper-connected online communities.
Shooglebox AI helps PR teams find and tune in to what people in these communities are doing, seeing and saying – and decode emerging trends and narratives.
The purpose-built AI tools are designed to:
-
Gather posts from TikTok, YouTube, Reddit and other platforms
​
-
Enhance each post with summaries, image analysis, video description and comment theming
​
-
Analyse everything and synthesise insights into internal reports of customer-facing content
​
-
Share boxes of at-a-glance contextualised examples of the posts, the conversations and the people behind them