Brand safety issues in advertising persist due to the industry’s reliance on “blunt, automated tools,” according to Nik Kontoulas, managing director of Seedtag ANZ.
Kontoulas said that existing solutions, such as keyword blocklists and domain exclusions, lack the nuance to truly serve advertisers and reward high-quality, ad-funded journalism.
“While brand safety is essential to protect brand reputation and avoid harmful associations, the way it’s operationalised, through automated blocklists and overly broad keyword exclusions, can have serious unintended consequences that limit reach and reduce campaign performance,” Kontoulas told B&T.
“Rather than relying on blunt tools, marketers need to embrace more nuanced, context-aware strategies that consider the actual meaning and sentiment of content. AI-powered contextual advertising delivers that level of precision. By understanding the tone and semantics of content, not just the presence of certain keywords, brands can ensure their ads appear in suitable, brand-safe environments without compromising scale.”
Seedtag offers marketers the peace of mind about where their ads are being placed through its contextual model that fully analyses the context and content of articles. Liz, Seedtag’s AI model, uses network-level embeddings to extract semantic, emotional and behavioural signals across multiple content formats such as text, imagery, video and audio.
“This allows Liz to interpret meaning holistically, going far beyond keyword or topic matching. Instead of simply placing ads near related content, Liz identifies where consumer engagement intersects with brand goals in real time,” said Kontoulas, adding that Liz operates “like a human brain,” being able to recognise patterns, interpret intent and respond dynamically to content.
“For example, it can understand the clear difference between a ‘terrorist attack’ and a ‘laugh attack.’ This deeper, more intelligent interpretation of content enables advertisers to scale campaigns more effectively, resulting in smarter, more unified strategies that deliver consistent and impactful outcomes across channels while maintaining alignment with brand values,” said Kontoulas.
This is not a purely academic exercise. There have been several stories in the industry this year surrounding brands unwittingly having their ads placed alongside harmful content. At best, this industry blindspot can lead to campaigns underperforming, while at worst, poor ad placements can end up financing and abetting some very serious crimes. Fortunately, it is easier for advertisers to avoid becoming entangled in these issues than to solve them entirely.
“By moving from rigid keyword blocklists to AI-powered contextual analysis, brands can target content that aligns with their values without penalising publishers for covering sensitive topics. This approach enhances brand safety and suitability by ensuring ads appear alongside relevant, value-aligned content,” said Kontoulas.
“AI-driven contextual technology also increases scalability, allowing campaigns to run broadly without losing precision. It supports healthier monetisation models for publishers by reducing blanket bans and enabling more targeted, meaningful ad placements. This alignment drives stronger engagement, better brand perception, and improved ROI for both brands and publishers.”
A Changing Media Market
The media market is changing, however. With consumers becoming used to on-demand, high-quality content from the major platforms and streaming providers, connected TV (CTV) is experiencing rapid growth, not least in Australia.
This change in consumer behaviour has also changed the way ads are served to them. Media buying on CTV is much more similar to buying on the open web, for instance, than traditional TV media buying. But this change does hold some problems for marketers, namely in the form of finding brand-suitable inventory.
“Brand suitability goes beyond traditional brand safety measures by ensuring ads appear in content that aligns with a brand’s values and objectives. It protects advertisers and publishers alike by semantically interpreting editorial content and understanding linguistic nuance. Our AI can assess content not just for safety but also for relevance, helping brands target meaningful and contextually appropriate content that resonates with their audience. This approach creates deeper engagement while maintaining safeguards against inappropriate or misaligned placements,” said Kontoulas.
“Contextual TV is revolutionising brand safety and suitability on CTV. Traditionally, brand safety focused on avoiding harmful or controversial content, but today’s advertisers want more: they seek brand suitability—content that aligns with their values and enhances their message.”
Again, Seedtag’s Liz AI is able to help. While the service isn’t live in ANZ yet, it has proven in other markets that it can effectively analyse video content at a scene level, assessing whether the tone, sentiment, and cultural relevance fit a brand’s identity.
“This capability is particularly important in CTV, where a single show may explore diverse themes. With AI-driven contextual understanding, advertisers can place their ads in content that not only avoids negative associations but enhances brand alignment and suitability,” said Kontoulas.
In fact, such is the breadth of Liz’s abilities, Kontoulas said Seedtag is entering a “new era” in contextual advertising.
“Traditionally, contextual has been limited to content recognition based on URL analysis, keywords, and categories—often limiting campaigns to the upper funnel and predefined audience segments. Our AI, Liz, changes this paradigm. Grounded in neuroscience principles and network-level embeddings, Liz mirrors how the human brain processes content—reflecting these dynamics and matching ad placements to moments of strong interest, emotional receptivity, and high intent. This evolution allows us to offer dynamic, custom audiences, 360° campaign insights, and full-funnel, omnichannel outcomes—from awareness to action.”
