The final day of July brings a clear message for the media and advertising industry: artificial intelligence is no longer sitting beside the existing digital ecosystem. It is becoming the ecosystem.
AI platforms are developing advertising products, search engines are replacing traditional links with generated answers, social platforms are turning discovery into direct commerce, agencies are rebuilding their operating models and publishers are trying to protect both their content and their audiences.
In Ireland, the industry is also seeing important leadership changes, continued growth in digital audio, new agency-built AI tools and fresh evidence that emotional connection still matters more than technology alone.
ChatGPT is becoming a serious advertising platform
One of the most important changes for advertisers is the continued development of advertising inside ChatGPT.
Businesses can now access ChatGPT advertising through agency and technology partners or through a beta self-service Ads Manager. Advertisers can set budgets, manage campaign pacing, upload creatives and review performance from within the platform.
ChatGPT Ads also supports both cost-per-thousand-impressions and cost-per-click buying. The introduction of CPC bidding moves the platform closer to the performance advertising models already used across search and social media.
Conversion tracking is being supported through pixel-based measurement and a Conversions API, allowing advertisers to measure purchases, leads, registrations and other actions after an ad interaction.
This is significant because ChatGPT conversations frequently take place while someone is researching, comparing products or deciding what to do next.
Traditional search advertising responds to a keyword. Conversational advertising can potentially respond to a much richer understanding of what a person is trying to achieve.
The challenge will be maintaining a clear boundary between independent AI answers and commercial messages. OpenAI says advertisements are displayed separately from answers, conversations remain private from advertisers and ad delivery does not influence the response produced by ChatGPT.
Brand safety could become ChatGPT Ads’ biggest differentiator
OpenAI has published detailed restrictions governing where advertisements can appear.
Ads are not intended to appear beside conversations involving areas such as political content, misinformation, graphic violence, mental health, personal crises, weapons, fraud or other sensitive situations.
The initial advertising categories are largely focused on consumer products, travel, education, digital services, household goods and local businesses. More heavily regulated industries are being introduced gradually or remain restricted.
For advertisers, this creates the possibility of a more controlled advertising environment than many open social or display platforms.
However, marketers will still need independent measurement. A conversational platform may generate fewer impressions than a large social network while producing users with much stronger intent. Judging the channel entirely through reach or click-through rates could therefore miss its real value.
The larger opportunity may be understanding whether ChatGPT influences consideration, shortlists and eventual purchasing decisions.
GPT-5.6 arrives as AI becomes part of everyday media production
OpenAI announced GPT-5.6 on 30 July, continuing the rapid development of models designed to combine stronger intelligence with improved efficiency. The company has also been expanding AI applications across research, workplace productivity, coding and news organisations.
For media and marketing teams, improvements in model efficiency matter almost as much as headline intelligence.
Faster and more affordable models make it practical to apply AI across everyday tasks such as research, reporting, campaign analysis, audience segmentation, creative adaptation, translation, social listening and content production.
The main question is shifting from whether businesses should use AI to where human approval remains essential.
Brands that automate everything risk producing large volumes of content that look and sound identical. The strongest operating model is more likely to combine machine speed with human judgement, cultural understanding and clear accountability.
Google introduces clearer labels for AI-generated advertising
Google is expanding the disclosure of generative AI use across advertisements appearing on Search, YouTube and Discover.
A new “How this ad was made” section within My Ad Center can indicate when an advertisement was created or altered using AI. Content made with Google’s own generative tools can be labelled automatically, while advertisers using external tools are being given controls to disclose their use of AI.
Local rules may also require a disclosure to appear directly on the advertisement.
The change reflects a wider industry movement towards AI transparency.
Consumers may not object to AI-assisted advertising, particularly when the technology is being used for localisation, resizing, translation or simple production work. The risk increases when a brand uses synthetic people, artificial testimonials or realistic scenes that could be mistaken for genuine events.
The label itself will not solve every issue. Advertisers will still need internal rules covering consent, likeness rights, factual accuracy, copyright, brand safety and the appropriate use of synthetic voices and faces.
Google Ads moves deeper into the Gemini era
Google’s advertising strategy is increasingly centred on Gemini-powered automation.
Its latest advertising products include conversational ad experiences, AI-generated shopping guidance, expanded creative tools and a unified adviser operating across Google Ads, Analytics, Merchant Center and Google Marketing Platform.
Google is also bringing AI Max into additional areas including Shopping and travel, while giving advertisers more control over brand tone, messaging and matching requirements.
Dynamic Search Ads are eventually expected to transition towards AI Max, although the automatic upgrade timeline has been extended and is now scheduled to begin in February 2027.
For advertisers, this creates both convenience and dependency.
Campaign creation, audience matching, bidding, creative production and measurement are increasingly being managed by the same platform. That can improve speed and performance, but it can also make it harder for advertisers to understand exactly why a campaign worked.
The future role of media teams will involve more than operating platform controls. They will need to test whether platform recommendations support the brand’s commercial strategy rather than simply increasing spend within the platform.
Meta AI moves from answering questions to completing tasks
Meta has expanded its AI assistant with more agent-like capabilities.
Meta AI can now complete research, prepare plans and presentations, deliver recurring briefings and respond to instructions while a task is still being completed. Meta says the assistant can combine information from the wider web with content and recommendations shared across Facebook, Instagram and its other platforms.
This could become highly valuable to Meta’s advertising business.
An assistant that helps someone plan a trip, furnish a home, find a restaurant or prepare for an event sits close to commercially valuable decisions. Marketplace listings, creator recommendations, business pages and paid placements could eventually become part of the same journey.
Meta has not simply created another chatbot. It is trying to connect AI with the vast amount of behavioural, commercial, creator and community activity already taking place across its platforms.
For advertisers, the concern will be whether Meta’s AI becomes another closed environment in which the platform controls discovery, recommendation, media delivery and measurement.
Meta’s image-generation controversy shows why consent matters
Meta also faced criticism over a feature that allowed public Instagram accounts to be referenced during AI image generation.
The company removed that capability after feedback, acknowledging that the feature had not provided the level of control people expected.
The incident provides an important lesson for brands and platforms.
Something being publicly visible does not automatically mean people expect it to be used as training material or as a reference for synthetic content. Legal permission, platform terms and public expectations are not always the same thing.
Brands using AI-generated images should therefore consider whether every recognisable person, creator, product and protected design has been used with proper permission.
TikTok wants to own the journey from discovery to purchase
TikTok is developing its advertising proposition around the idea that product discovery, culture and commerce now happen together.
Its latest advertiser tools use AI to identify customer intent, improve creative development, personalise marketing and connect content discovery with measurable business outcomes.
TikTok has also introduced performance-focused advertising products for entertainment and streaming businesses in Europe. Streaming Ads can promote relevant programmes from a platform’s catalogue based on demonstrated user interests, while New Title Launch is designed around major entertainment releases.
The platform’s growing position in sports and entertainment was visible during the 2026 football tournament, where related content generated more than one trillion views. TikTok has also announced partnerships spanning basketball, music releases and major live performances.
This makes TikTok increasingly difficult to classify as only a social platform.
It now operates as a search engine, entertainment service, creator marketplace, product discovery tool, live-event destination and advertising platform.
AI search is creating a traffic emergency for publishers
While technology platforms expand their AI products, publishers are facing a much more difficult reality.
Publisher advertising supply reportedly fell by as much as 40% during the second quarter as AI-generated search answers and zero-click experiences reduced the number of people visiting external websites.
Academic research has also found evidence that exposure to Google AI Overviews can reduce visits to informational pages. One large study involving matched Wikipedia pages estimated an approximately 15% reduction in daily traffic for exposed English-language articles.
Another study found that AI Overviews appeared far more frequently for question-based searches and that some of the claims included in generated summaries were not fully supported by the pages being cited.
This creates an uncomfortable situation for media owners.
Publishers pay journalists, creators and experts to produce information. AI systems then summarise that material directly on a platform, reducing the need for the user to visit the original publisher.
The platform retains the audience and advertising opportunity, while the publisher loses the page view.
Media businesses need audiences they control
The decline of referral traffic makes owned audiences increasingly valuable.
Email newsletters, mobile applications, memberships, events, subscriptions, podcasts and direct communities give publishers a relationship that cannot be removed by a search algorithm change.
Search optimisation will remain important, but publishers must also think about visibility inside AI-generated answers. This includes producing authoritative original reporting, using clear page structures, demonstrating subject expertise and making information easy for AI systems to understand and cite.
Being cited by an AI platform, however, will only be valuable when that citation produces recognition, trust, subscriptions or meaningful referral traffic.
Publishers should avoid treating AI visibility as a replacement for actual audience ownership.
AI crawlers are becoming a commercial negotiation
Publishers are also becoming more selective about which automated systems can access their websites.
Cloudflare is preparing controls that categorise AI traffic as search, agent or training activity. The intention is to give website owners more control over whether AI systems can index content, perform tasks or use information for model development.
The risk is that blocking the wrong crawler could also affect search visibility, particularly when the same bot performs several functions.
This is likely to become one of the largest commercial disputes in digital media.
Platforms need high-quality information to produce useful AI answers. Publishers need compensation, attribution and traffic to finance the creation of that information.
The future relationship may involve licensing agreements, paid access, crawler-level permissions and technical standards that distinguish between search indexing, AI answers, training and autonomous agents.
Small creators are becoming more valuable to large brands
Brands are increasingly working with ordinary customers and small creators rather than relying only on celebrities or accounts with millions of followers.
Some creator programmes now accept people with only a few hundred followers, rewarding them with products, discounts or modest payments in exchange for authentic content.
The approach reflects how social algorithms have changed. A post can reach a significant audience based on engagement and relevance even when the creator has a relatively small existing following.
At the other end of the scale, Unilever reportedly activated around 50,000 creators across 120 markets during the 2026 football tournament as part of a much larger creator network.
The advantage is volume, localisation and cultural relevance.
The risk is losing control over disclosure, brand consistency and creative quality. Managing thousands of creators requires stronger briefing systems, content approval processes, rights management and measurement standards.
Creator marketing is becoming an operating system rather than a collection of individual influencer campaigns.
Irish digital audio continues to grow
Ireland’s radio advertising market generated approximately €79.6 million during the first six months of 2026.
Overall revenue increased only slightly, but digital audio grew by 23.1% to reach €4.4 million. Branded content revenue also increased by 5% to €15.2 million, while traditional spot advertising declined by 2.1%.
These figures suggest advertisers are changing how they use audio.
Instead of relying entirely on standard radio commercials, brands are spending more on sponsorships, partnerships, podcasts, digital audio and content designed around particular presenters or audiences.
Audio also benefits from being available across several environments, including traditional radio, phones, connected cars, smart speakers and podcast applications.
For brands seeking both reach and trusted editorial environments, Irish audio remains one of the more interesting areas of media growth.
AI-built creative tools emerge inside Ireland’s agency market
Irish agencies are beginning to develop their own AI products rather than depending entirely on tools supplied by global platforms.
Wolfgang Digital has introduced two AI-powered systems designed to connect audience intelligence with short-form video performance.
One tool analyses conversations, sentiment and recurring themes across social platforms and customer-review sources. The second monitors how individual creative assets, formats and messages perform after a campaign launches.
The agency says the tools have contributed to a 65% improvement in short-form video hold rates and double-digit revenue growth for clients.
This represents a practical use of AI.
Instead of using the technology merely to create more images or videos, the tools are designed to understand what audiences are discussing and feed performance results back into the creative process.
The agencies that succeed with AI may be those that build proprietary intelligence around clients and consumers rather than using the same general tools available to everyone else.
Major agency groups place AI at the centre of growth
The financial results of global agency groups show that AI is becoming central to how they describe their future.
Omnicom reported second-quarter core revenue of approximately $6 billion, representing organic growth of 6.1%. Integrated media generated more than half of core revenue.
The company identified agentic marketing, social and creator activity, connected commerce, sports, entertainment and AI-driven discovery as important areas of future growth.
Publicis reported second-quarter organic net revenue growth of 4.8%. The group said AI-powered marketing services represented 87% of net revenue and grew organically by 6.5% during the quarter.
These numbers do not mean that 87% of advertising is being created independently by machines.
They show how broadly agency groups now apply the term AI-powered across data, media, creative, commerce, customer relationship management and production.
The holding-company competition is no longer only about agency networks or media buying scale. It is increasingly about who owns the strongest data, identity systems, technology partnerships and AI infrastructure.
Leadership changes reshape Irish media
The Irish media industry is also experiencing important senior leadership changes.
Bill Kinlay is stepping down from WPP Media at the end of August, with Ken Nolan appointed interim chief executive. Kinlay has spent almost four decades within the organisation and played a central role in establishing Mindshare in Ireland.
Karen Preston is also preparing to leave her role as chief commercial officer of Mediahuis Ireland in October.
Her time with the business included the development of a more diversified digital and multi-platform commercial model, including new approaches based on subscriptions, first-party audience intent and digital revenue.
These transitions arrive at a time when Irish media companies must manage falling search traffic, changing subscription behaviour, AI licensing questions, new advertising formats and growing competition from global platforms.
Tayto remains Ireland’s most emotionally connected brand
Despite the industry’s focus on automation, data and artificial intelligence, Ireland’s latest brand rankings provide a reminder that emotional connection remains difficult to automate.
Tayto retained first place in the 2026 Brand Reaction Index, followed by Penneys, Dunnes Stores and Lidl.
The Wild Atlantic Way, Netflix, the Credit Union, An Post and Brennans Bread also appeared in the top ten.
The strongest-performing brands tend to combine everyday presence with trust, cultural meaning, consistency and distinctiveness.
Technology can improve targeting, production and measurement, but it cannot automatically make a brand meaningful.
A highly personalised advertisement is not necessarily memorable. A perfectly automated campaign cannot compensate for a brand that has no recognisable personality or place in people’s lives.
The real media trend is the collapse of separate channels
The biggest development of 2026 is not one individual AI model or advertising product.
It is the disappearance of the boundaries between search, social media, commerce, entertainment, customer service and advertising.
ChatGPT is becoming an advertising and product-discovery environment. Google is combining generated answers with personalised commercial experiences. TikTok is connecting culture with shopping. Meta is building an assistant using content from its social and commercial ecosystem.
At the same time, agencies are becoming technology operators, creators are becoming media networks and publishers are being forced to operate as membership, events, audio, subscription and data businesses.
The industry is moving from a collection of digital channels towards a smaller number of intelligent platforms that attempt to manage the complete customer journey.
For advertisers, this creates extraordinary efficiency but also greater platform dependency.
For media businesses, it makes direct audience relationships essential.
For brands, it makes trust, distinctive creative thinking and cultural relevance more valuable—not less—as automated content becomes easier for everyone to produce.
