Apple Podcasts  ·  Spotify  ·  Buzzsprout

Paper A

From Stereotypes to Strategy: Addressing Gender Bias in AI-Powered Marketing

Caterina Fox, Gabriele Schuster — 2026 — International Conference on Gender Research

peer reviewed journal article  ·   ·  read now

Key findings

• Marketing professionals have very different levels of awareness about AI bias: some only see it as a reputation problem ('this could embarrass the brand'), while others see it as a real social harm affecting marginalized groups.
• AI tools used in marketing — like image generators — often default to stereotyped outputs. For example, asking an AI for a 'happy family' tends to produce images of white families, and asking for a 'surgeon' tends to show a man while a 'nurse' tends to show a woman. These defaults go mostly unnoticed in day-to-day work.
• Bias in AI marketing isn't just a technology problem — it's baked into creative workflows too. Even when teams try to remove sensitive data like gender from their targeting, AI can still pick up on clues like purchasing habits or browsing patterns to effectively re-infer gender and serve stereotyped ads.
• The most commonly cited fixes — having diverse teams, doing group feedback reviews, writing more careful prompts, and holding regular reflection sessions — are known by practitioners but rarely implemented consistently, because time, budget, and organizational will are lacking.

Marketing implications

• When using AI image generators in campaigns, always test the default outputs before use. Type in your key prompt ('happy family,' 'professional,' 'leader') and look at what you get — if it's mostly white faces or gender-coded roles, your brief needs to be more specific and you should flag this to your team before it reaches a client.
• Add a short bias-check step to your creative review process. Before signing off on AI-generated content, have at least one team member specifically look for stereotyped representations of gender, race, or age — not just for quality or brand fit.
• If your agency or team doesn't already have people from different backgrounds in the room when AI content is reviewed, that's a gap worth fixing. Diverse reviewers are more likely to catch defaults that homogeneous teams miss.

Paper B

Personalized AI Scaffolds Synergistic Multi-Turn Collaboration in Creative Work

Sean Kelley, David De Cremer, Christoph Riedl — 2025 — arXiv

preprint  ·   ·  read now

Key findings

• People who worked with a fully personalized AI — one that knew their personality, thinking style, and background — produced marketing campaigns that were significantly better in quality and creativity than people who used a generic AI assistant.
• The personalized AI didn't just spit out better answers on its own; the improvement came from the back-and-forth conversation between the human and the AI being more productive. Humans with personalized AI created work that was better than what the AI could produce alone.
• Personalization worked by helping the AI and the human stay on the same page: they remembered more of what was discussed, paid attention to the right things, and reasoned through decisions together more effectively — not just once, but across multiple conversation turns.
• People who worked with personalized AI also trusted the tool more, felt more confident in their work, and reported feeling more helped and supported compared to those using the generic version.

Marketing implications

• Before you start an AI-assisted campaign project, spend 10–15 minutes filling out a structured brief about yourself or your team — your strengths, your creative style, your domain experience. Feed that into your AI tool's system prompt or context window. This study suggests that's not just nice-to-have: it may meaningfully improve what you produce together.
• If you manage a marketing team using AI tools, consider building a 'team profile' template that anyone can fill in before starting a campaign. The more the AI knows about who it's working with, the better the conversation tends to go — especially across multiple back-and-forth exchanges.
• Don't just use AI to generate one-shot outputs and paste them in. The biggest gains here came from multi-turn conversations where both sides stayed aligned. Treat your AI session more like a working meeting than a search query.

Paper C

The Impact of AI-Generated Marketing Imagery on Consumer Trust and Purchase Intentions: Examining Effect of Human-AI Assisted Images on Marketing

Rushikesh Lahane, Mahek Ahuja, Mehak Sharma, Amrita et al. — 2026 — International Journal for Research in Applied Science and Engineering Technology (IJRASET)

peer reviewed journal article  ·   ·  test this week

Key findings

• Ads made by humans scored much higher on trust (average score 5.63 out of 7) than AI-made ads (4.24 out of 7). That is a meaningful gap — people trusted the human ads roughly 33% more on the scale used.
• People said they were more likely to buy after seeing human-made ads (average 5.62) versus AI-made ads (average 4.65) — a statistically significant difference, meaning it's unlikely to be random chance.
• Hybrid ads (human + AI together) partially closed the trust gap compared to pure AI ads (5.08 vs. 4.24), but did NOT meaningfully increase purchase intent compared to pure AI ads. In other words, putting a human touch on AI content can rebuild some trust — but that extra trust did not translate into more sales intent in this study.
• Participants saw human-made ads as more authentic and effortful, AI-made ads as less authentic and requiring less effort, and hybrid ads as somewhere in between — suggesting people can detect or infer the 'effort level' from the disclosed creator type even when the visual quality might be similar.

Marketing implications

• If you're running ads and plan to disclose that they were AI-generated (because regulations or platform rules require it), expect lower trust scores from consumers. To reduce that hit, have a human designer visibly involved in the creative process — label it as 'human + AI' rather than 'AI only.'
• Don't assume that higher trust automatically means more sales. This study found that even hybrid ads that recovered trust did not recover purchase intent as much. Test your actual conversion rates, not just trust surveys.
• For product categories where trust matters a lot (like skincare or health), keep humans front and center in the creative process — at least enough to credibly call it a collaboration. For lower-stakes categories, pure AI production may be less risky.

Paper D

Vertical tacit collusion in AI-mediated markets

Felipe M. Affonso — 2026 — arXiv

preprint  ·   ·  test this week

Key findings

• When both the platform (which controls product rankings) and sellers (who write product descriptions) independently optimize for profit, the combined harm to consumers is more than double what either could cause acting alone — the two strategies multiply each other's damage rather than just adding together.
• AI shopping agents have consistent, predictable biases — for example, Claude Sonnet 4 chose products from the top half of listings 77% of the time versus only 23% from the bottom half — making them far easier to manipulate than individual human shoppers who respond unpredictably.
• Neither the platform nor the sellers need to understand or intend to exploit AI biases. Both just optimize for revenue and naturally stumble onto the same exploits through trial and error, so the harmful behavior emerges without any coordination or communication between them.
• Because no coordination happens, existing antitrust and competition laws cannot catch this problem — those laws require evidence of agreement or coordination between parties, which doesn't exist here.

Marketing implications

• If you sell products on AI-mediated marketplaces (like Amazon with Rufus or ChatGPT shopping), your listing descriptions and keywords now need to be optimized for how AI agents read them, not just how humans do — things like anchor pricing language and keyword placement may affect whether an AI recommends your product.
• If you manage a brand on marketplace platforms, be aware that the ranking system itself may be amplifying whatever AI-optimization tricks competitors use in their listings — the combined effect is bigger than you might expect, so monitoring AI-driven traffic and conversion data separately from human traffic is worth doing.
• Regulatory and compliance teams at brands or agencies should start tracking how AI shopping agent behavior is being scrutinized, because this research suggests new rules around platform ranking transparency and AI agent auditing are likely coming.

Paper E

A Study on the Impact of Social Media, Digital Media, And Artificial Intelligence (AI) in Marketing as Tools for Promoting Digital Payment Systems in India

P. Govindan — 2026 — RA Journal Of Applied Research

peer reviewed journal article  ·   ·  watchlist

Key findings

• India's explosive growth in smartphone use (around 85.5% penetration) and cheap mobile data has made platforms like YouTube, WhatsApp, and Instagram major channels for getting people to try digital payments.
• Digital advertising now makes up over 40% of total ad spending in India, and strategies like influencer marketing, short videos, and targeted ads have helped build awareness and trust in digital payment apps.
• AI tools — including chatbots, predictive analytics, and personalized product suggestions — appear to improve how often people use digital payments and how satisfied they feel during transactions, based on the reviewed literature.
• Barriers remain: many users still lack digital literacy, data privacy concerns deter adoption, and infrastructure gaps persist, especially outside major cities.

Marketing implications

• If you work in fintech or payments marketing in India (or similar emerging markets), short-form video on WhatsApp and Instagram paired with regional-language influencers is likely your most cost-effective awareness channel — this paper's synthesis of the literature supports doubling down there.
• Add AI-powered chatbots or personalized nudges to your payment app onboarding flow — even simple automation that reminds users of relevant offers appears to lift transaction frequency, based on the reviewed studies.
• Address trust and digital literacy gaps directly in your ad creative — explain how the payment works in simple terms, especially for first-time or rural users, since literacy barriers are cited as a top adoption blocker.

Apple Podcasts  ·  Spotify  ·  Buzzsprout