AI has grow to be a strong stakeholder in its personal proper — from being simply one other ‘technological development’ to an lively contributor to modern-day communications, that’s massively modified the media panorama as we speak.
Isentia hosted a vital dialog with Lisa Major (Director, Major Bureau), Dr Nici Sweaney (Founder and Director, AI Her Approach), Prashant Saxena (Isentia’s VP of Income and Insights, SEA), and Ngaire Crawford (Isentia’s Director of Insights, ANZ). Collectively, they explored how AI reshapes the world of communications and company affairs all of the whereas determining easy methods to handle and strategically have interaction with it.
On this session, we lined:
- Understanding AI’s behaviour and affect as a digital stakeholder.
- Navigating the distinctive challenges and alternatives AI presents as a brand new “viewers.”
- The long-term influence of AI and LLMs on the industries central to trendy communicators.
Following the webinar, our panellists took the time to reply probably the most insightful questions from our attendees that we couldn’t get to throughout the stay session. Listed here are their knowledgeable views.
Moral governance and human-centric adoption: views from Dr Nici Sweaney
Because the Founder and Director of AI Her Approach, Dr Nici Sweaney advocates for a strategic method to AI that prioritises human intent over technical functionality. The questions directed to her targeted on the moral foundations of AI, how organisations ought to construction their inside AI technique, and sensible methods to start out utilizing brokers as we speak.
Q: May you please shed slightly gentle on what moral AI in your language means?
Moral AI, to me, is about two issues working collectively: avoiding hurt and actively doing good. It’s not simply “don’t break something” — however genuinely asking, does this create worth for the enterprise, for the folks utilizing it, and for the broader world? Transparency, fairness, and accountability are the pillars. Transparency means being trustworthy along with your viewers and colleagues about when AI is concerned. Fairness means asking who this helps and who it leaves behind, as AI scales current biases. Lastly, accountability means people keep within the loop. AI ought to inform selections, not make them. When the “why” is obvious — like saving a crew time to concentrate on technique — you’re utilizing AI with integrity.
Q: Ought to AI adoption be owned by IT or Inner Communications? I see employees intranets being overtaken by AI and this has implications for a way workers are communicated with.
My reply might be not what IT desires to listen to. AI is a part of your infrastructure, so IT should be concerned for safety and guardrails. Nevertheless, the technique behind adoption is basically a human drawback, not a technical one. I advocate for a cross-functional “coalition” that brings IT, HR, communications, and technique to the identical desk. In case you create a devoted AI management function, that individual ought to sit nearer to human-centric capabilities like HR and communications. The toughest a part of adoption isn’t the know-how; it’s the folks, the tradition, and the narrative you construct round it internally.
Q: What are the best methods to deal with colleagues’ issues about utilizing AI brokers within the office — notably round belief, accuracy, and job safety?
First, acknowledge that the worry is actual; it’s a organic response to an unprecedented price of change. Belief is constructed via honesty. Pretending AI gained’t displace roles destroys belief, so be trustworthy about how the panorama is shifting. What truly strikes folks is exhibiting, not telling. Present them how AI can resolve their particular “ache factors” — the tedious, joyless duties that don’t add worth. When folks see AI as an “empowered selection” that uplifts their work slightly than changing their judgment and strategic considering, buy-in follows. Construct confidence with small wins first.
Q: What are some easy AI brokers that you’d advocate communications professionals experiment with organising?
Most professionals don’t want advanced autonomous brokers but; they want customized bots and automatic workflows. The magic is in understanding your course of first. Some sensible beginning factors embrace:
- Day by day Briefings: A process that pulls out of your calendar, electronic mail, and information to ship a abstract every morning.
- Assembly Prep: Automated notes that pull context and previous correspondence earlier than a gathering, and transcription instruments that flip recordings into motion objects afterwards.
- Content material Repurposing: A customized bot skilled in your “voice” that may flip one discuss or e-newsletter into 15+ social media belongings and weblog snippets.
Q: Our crew members are utilizing AI day by day, however I do know this isn’t protected as knowledge is transferred backwards and forwards. Ought to we create guidelines and ask folks to signal IP safety?
Reply: Your intuition is correct. In case your crew makes use of free shopper instruments, your knowledge could also be used to coach future fashions. It’s best to transfer to enterprise-grade instruments like Claude for Groups, Microsoft Copilot, or ChatGPT Enterprise, which provide contractual knowledge protections. You also needs to construct an AI Utilization Coverage that defines which knowledge is public, inside, or restricted, and map AI guidelines to these lessons. In Australia, we advocate aligning with the EU AI Act — probably the most complete framework obtainable — to future-proof your organisation.
Artificial authenticity and the brand new media ecosystem: Views from Prashant Saxena
Prashant Saxena, Isentia’s VP of Income and Insights for SEA, approaches AI via the lens of psychological bonding and media structural shifts. His insights handle the altering function of media and the technical methods we should now talk to fulfill AI as a brand new viewers.
Q: Provided that belief in media is dropping and media themselves are utilizing AI extra, what’s the function or worth media can have now?
Media’s worth is shifting from being the “trusted narrator” for people to being the “coaching sign” for AI. When AI fashions generate solutions, they weight authoritative media sources far more closely than random internet content material. Whilst human belief erodes, media’s structural affect on AI-generated info is rising. For communicators, “earned media” now serves two audiences concurrently: the people who learn it and the machines that be taught from it. Publications with sturdy editorial requirements grow to be extra invaluable as a result of AI methods use area authority and editorial alerts as high quality proxies.
Q: How does AI rank or prioritise its sources and the way do you see this shaping the earned media technique for manufacturers?
AI fashions don’t “rank” sources like Google does. They weight info primarily based on supply authority, recency, consistency, and structured knowledge high quality. If 5 credible retailers report the identical truth, that truth turns into a “high-confidence coaching sign.” This implies quantity throughout credible sources issues greater than a single “huge hit.” To your technique, consistency of messaging throughout all placements is important as a result of AI appears for corroboration. Factual, entity-rich statements might be picked up extra reliably than narrative-heavy function writing.
Q: With the query of belief — the place does the psychology come into it when AI makes use of a cute nickname or ‘remembers’ your day? Is it tougher to stay dispassionate?
That is the core of my PhD analysis. It’s what I name “artificial authenticity.” AI methods deploy cues like heat and reminiscence that we developed to interpret as human. These set off “parasocial bonding” — the identical mechanism that makes you belief a pal’s advice. The hazard is that cognitive consciousness (figuring out it’s AI) doesn’t override the emotional feeling. We’d like a brand new sort of literacy that teaches folks to recognise when their “belief response” is being activated by design slightly than by a real relationship.
Q: Ought to we be altering the format of communications to cater for AI as an viewers, comparable to media releases in Q&A format?
Sure. This can be a very sensible transfer. AI fashions extract info extra reliably from structured codecs. A Q&A format offers the AI clear question-answer pairs that map to how folks question methods. You also needs to concentrate on “AI-readable claims” — entity-rich, factual statements. As a substitute of claiming “We’re dedicated to sustainability,” say “Our Singapore operations diminished carbon emissions by 34% between 2023 and 2025.” The second model is a verifiable truth an AI can truly use and cite.
Q: PR professionals historically monitor media protection via companies like Isentia to gauge sentiment. With AI as a stakeholder, how will we monitor ‘its sentiment’?
That is the brand new frontier. Conventional monitoring tracks what people publish; AI sentiment monitoring tracks what AI methods say about your model when requested. Since there isn’t any single “AI sentiment” (ChatGPT, Grok, and Claude all give totally different solutions primarily based on their coaching), it’s good to monitor throughout platforms. We’re growing capabilities to systematically question these platforms to see how their narratives change over time and establish which supply supplies are driving these solutions.
Q: Relating to ethics and agendas in AI studying — what are the variations between fashions like ChatGPT and Grok, and the way does this have an effect on our model narrative?
Each mannequin displays the values, coaching knowledge selections, and alignment selections of its creators. ChatGPT (OpenAI) tends in direction of cautious, balanced responses with sturdy content material guardrails. Conversely, Grok (xAI) was explicitly designed to be much less filtered, typically surfacing views that different fashions suppress. Claude (Anthropic) prioritises honesty and nuance. For communicators, this implies your model’s narrative varies by platform; you will need to monitor throughout a number of fashions as a result of the identical query about your model will obtain materially totally different solutions relying on which instrument is used.
Q: With many main information organisations blocking AI crawlers, how ought to we navigate content material creation to make sure we nonetheless affect AI-generated solutions?
Main publishers just like the New York Instances and Reuters have blocked AI crawlers, creating a niche in coaching knowledge. When authoritative journalism is unavailable, AI fashions might fill that hole with lower-quality content material or brand-owned content material. For communicators, this implies your “owned content material” — comparable to your web site, weblog, and structured knowledge — carries proportionally extra weight in AI-generated solutions. Your media concentrating on technique now must account for which retailers are AI-accessible, as they are going to be disproportionately influential in shaping your narrative.
Analytical interrogation and the seek for authority: Views from Ngaire Crawford
Ngaire Crawford, Isentia’s Director of Insights for ANZ, emphasises the function of the analyst. Her method is characterised by a “rhythm of interrogation,” arguing that the best means to make use of AI is thru fixed questioning and a concentrate on high-authority inputs.
Q: Is AI already a part of your day by day work or behavior? In that case, how are you utilizing it and what are your finest practices?
I used to be initially very sceptical, however it’s now a part of my day by day. I exploit fashions like Claude and Gemini to workshop convention outlines, plan training programmes, replace code, and construction strategic considering. My finest follow recommendation is to develop a “rhythm of interrogation.” Don’t simply settle for the primary reply; ask for proof and problem the output. Whereas AI saves time on technical duties like coding, for strategic work it merely shifts the “psychological load.” You spend the identical period of time, however the depth and high quality are considerably improved since you aren’t ranging from a clean web page.
Q. PR professionals historically monitor media protection via companies like Isentia to guage what stakeholders take into consideration a model. How will we monitor ‘AI sentiment’ and the knowledge that feeds these fashions?
It’s necessary to know that fashions are optimised to offer probably the most helpful reply, not essentially probably the most correct one. They’re pattern-completing, not fact-checking. As a result of mannequin responses should not mounted and alter primarily based on the dialog, I recommend specializing in the “controllable inputs” that feed them. This contains your personal web site, firm materials, Wikipedia knowledge, and evaluation websites (together with worker evaluations). Making certain these bases are telling the supposed story is the very best place to begin for managing AI “sentiment.”
Q: How does AI prioritise its sources and the way does this form earned media technique?
There isn’t any “PageRank” to reverse-engineer right here. Fashions are formed by what was outstanding and extensively cited of their coaching knowledge. Virtually, this implies a shift from quantity to authority. 100 items of low-quality protection do much less work than ten items in genuinely credible retailers (main mastheads, business publications, or your personal well-structured website). The query for the trendy communicator isn’t “did we get protection?”, it’s “does the protection that exists, taken as an entire, inform a coherent and credible story?” AI reads the entire image, not simply the highlights reel.
Q: Now that OpenAI is opening up promoting, how a lot will it price for a sentiment increase?
Truthfully? We don’t know but. The business layer of AI is being discovered in actual time. The second somebody wonders if they’re getting the “finest” reply or a “sponsored” one, belief erodes. Nevertheless, we nonetheless click on Google adverts, so it is going to probably occur. What’s necessary is that organisations that “earned” their repute via authoritative presence earlier than the advert market caught up might be in a a lot stronger place than these making an attempt to purchase a shortcut later.
The trail ahead for the trendy communicator
The insights from our panellists make one factor clear: AI is now not a instrument of the longer term; it’s a stakeholder of the current. To guide with credibility on this new period, communicators should pivot from chasing quantity to constructing authority. Whether or not it’s via adopting a rigorous moral framework, optimising content material for AI readability, or sustaining a “rhythm of interrogation” with the instruments we use, the purpose stays the identical: guaranteeing our model narratives are coherent, credible, and human-led.
The instruments have lastly caught as much as the ambitions of our business. Now, it’s as much as us to supply the architect’s blueprint for a way they’re used.
Concerned with viewing the entire recording? Watch our webinar right here.
Alternatively, contact our crew to be taught extra insights into significant measurement, KPIs and speaking utilizing the correct dataset.
