What AI Can and Cannot Do in Ad Creative: Expectations vs. Reality

Meta description: AI-powered creative automation tools for display advertising promise faster HTML5 banner production, scalable ad variations, and reduced burnout. This guide explains what AI can and cannot do in ad creative, compares creative automation platforms, and shows how infrastructure like Viewst helps teams automate HTML5 banner creation without sacrificing brand control.

Slug: creative-automation-tools-display-advertising-platform-comparison

If you're evaluating the best software for creating and managing display ads, you've likely seen bold claims about “AI doing all your creative.”

In reality, the most effective creative automation platforms for display advertising treat AI as production infrastructure—not as a replacement for designers.

This article breaks down:

• What AI actually does well in banner ad production

• Where human creativity and judgment remain indispensable

• How tools like Viewst fit between design and media as production infrastructure

• A practical platform comparison and features overview for creative automation tools

What AI Does Well in Display Ad Production: Creative Automation Platforms & Banner Ad Production Automation

Across the industry, AI is already a proven production multiplier.

Adobe’s 2024 global survey of 2,727 creative professionals found:

• 83% were already using generative AI

• 62% said AI reduced task time by about 20%

• 66% felt they were making better content

• 58% reported producing more content overall
  (Source: Scott Belsky et al., “Creative Pros and Generative AI Usage,” Adobe, Feb 2, 2024, https://blog.adobe.com/en/publish/2024/02/02/creative-pros-generative-ai-usage)

In Adobe’s 2026 follow-up study (4,000+ creatives across US, UK, and global markets):

• Creatives used AI on more than 40% of projects

• Nearly 9 in 10 said AI had improved their work
  (Source: Adobe Research, “How Creatives Are Thinking About AI,” Adobe, Apr 17, 2026, https://blog.adobe.com/en/publish/2026/04/17/creatives-say-ai-helping-them-meet-growing-demand-content-improving-their-work)

Within display ad production, AI excels at:

• Automating HTML5 banner creation from a master creative
  Generating size variants and structural layouts.

• Automated asset resizing for display ads
  Smart resize tools convert a single design into dozens of IAB-standard formats.

• Scalable ad variations with AI
  Copy swaps, image variations, and localized versions for multi-market campaigns.

• Routine production tasks
  Exports, file naming, motion presets, and QA on specs.

This is where tools like Viewst deliberately focus.

Viewst’s capabilities are vendor-provided product features (as described in Viewst documentation and marketing materials):

• AI Smart Resize — from one master HTML5 creative, generate all required display sizes.

• AI Image Deflatening — turn flat image assets into editable, layered HTML5 designs.

• AI Designer — convert prompts and brand rules into structured HTML5 banner sets.

• AI Instant Animator — apply motion to layers with one click.

These features are designed for banner ad production automation, not for fully autonomous concept creation.

Where Human Creativity Remains Indispensable (With Evidence)

Claims that “AI will replace designers” are not supported by current research.

A 2024 design study on GenAI-assisted design (global sample; controlled lab evaluation) found that designs supported by generative AI were judged:

• More creative and unconventional

• But not significantly better on visual appeal, brand alignment, or usefulness
  (Source: Jong-Yun Kim et al., “Generative AI in Visual Design: Creativity vs. Brand Fit,” arXiv preprint 2411.00168, 2024, https://arxiv.org/abs/2411.00168)

This suggests:

• AI can push novelty and variation.

• It does not reliably guarantee brand fit or usefulness without human oversight.

We should therefore say:

• AI often reduces mechanical work and supplies options.

• Human teams remain necessary for brand judgment, strategy, and final approval.

In display ad production, humans still own:

• Brand and strategic direction
  Positioning, messaging hierarchy, and campaign concepts.

• Tone, narrative, and empathy
  Understanding audience context, sensitivities, and cultural nuance.

• Creative accountability
  Ensuring work aligns with brandbook, performance goals, legal rules, and ethical standards.

• Final craft decisions
  Composing layouts, nuanced typography choices, and the difference between “good enough” and “on-brand.”

Adobe’s research aligns here: creatives use AI most in brainstorming, research and ideation, but not in client interaction or final decision-making.
(Source: Adobe Research, “How Creatives Are Thinking About AI,” Adobe, Apr 17, 2026, https://www.adobe.com/ai/research/202606/how-creatives-are-thinking-about-ai.html)

The Real Problem: Volume, Repetition, and Designer Burnout

For most teams, the bottleneck is production volume, not a lack of ideas.

Adobe Express’s 2023 survey of 1,010 US business owners and marketers reported:

• 21% frequently feel burned out by content demands

• 33% doubled content output in the previous year

• AI-using teams produced 75% more content per week than non-users

• AI users saved an average of 14 hours per week

• 44% used AI specifically to uphold brand voice and standards
  (Source: Adobe Express, “How to Scale Content Creation with AI,” Adobe, Aug 2023, https://www.adobe.com/express/learn/blog/scale-content-with-ai)

Earlier Adobe “State of Creativity” research found 44% of creatives spend half their week on repetitive work.
(Source: Adobe, “State of Creativity,” Adobe, 2022, https://www.adobe.com/creativecloud/design/discover/state-of-creativity.html)

In display advertising, that repetition looks like:

• Resizing banners into 15–30 formats per campaign

• Producing local-market versions for multiple languages

• Running performance variants for ongoing A/B testing

• Managing exports, spec checks, and manual QA for each ad server

This is precisely what Viewst is built to absorb:

• Infrastructure for banner production
  One place to scale formats, apply animation, manage brandbooks, and export ad-ready HTML5/GIF/MP4.

• Respect for expertise
  Senior designers stay focused on creative direction while production infrastructure handles non-creative suffering.

AI Adoption vs. Governance: Why Brand Control Matters

AI adoption in marketing is already high, but governance often lags.

The IAB (Interactive Advertising Bureau) 2025 State of Data report surveyed 223 agencies, brands, and publishers across the US and global markets:

• Only 30% had fully integrated AI across the media campaign lifecycle

• Roughly half of the rest expected to do so by 2026
  (Source: David Cohen, “IAB State of Data 2025,” IAB, Feb 2025, https://www.iab.com/news/iab-state-of-data-report-2025/)

IAB’s 2025 responsible-AI research (US marketers, n ≈ 300) found:

• Over 70% had experienced at least one AI-related incident (hallucination, bias, off-brand content) in advertising
  (Source: IAB, “AI Adoption Is Surging in Advertising, But Is the Industry Prepared for Responsible AI?” IAB Insights, Oct 2025, https://www.iab.com/insights/ai-adoption-is-surging-in-advertising-but-is-the-industry-prepared-for-responsible-ai/)

And their 2026 transparency study, covering US consumers (n ≈ 2,000) and ad executives, found:

• 82% of ad executives believed Gen Z/Millennials feel positively about AI-generated ads

• Only 45% of consumers actually expressed positive sentiment

• 73% of Gen Z and Millennials said clear disclosure of AI use would increase or not affect purchase likelihood
  (Source: Caroline Giegerich, “AI Transparency and Disclosure Framework,” IAB, May 2026, https://www.iab.com/news/iab-releases-industrys-first-ai-transparency-and-disclosure-framework-to-guide-responsible-advertising-in-a-generative-ai-landscape/)

As IAB’s VP of AI Caroline Giegerich notes, disclosure should be material-based: teams should disclose when AI meaningfully affects authenticity, identity, or representation.

For creative operations, this means:

• AI must be paired with brandbooks and governance.

• Automation should reinforce brand standards, not bypass them.

• Review and approval flows should live alongside the creative—inside the production environment, not scattered in email threads.

Viewst’s brandbook and governance features are vendor-described capabilities:

• Locked typography, colors, and component behavior

• Brandbooks applied across all sizes and variations

• Collaborative commenting and approvals inside the banner set

Dynamic Creative Optimization vs. Human Creativity

Dynamic Creative Optimization (DCO) platforms (Amazon, Google, Yahoo, etc.) show how AI excels at combination and optimization, but they assume strong human creative foundations.

For example:

• Amazon Ads allows advertisers to upload up to 15 images and 15 headlines for Responsive e-Commerce creatives, with machine learning optimizing combinations at scale.
  (Source: Amazon Ads, “Personalize Ads with Amazon DSP Responsive e-Commerce Creatives,” Amazon, 2024, https://advertising.amazon.com/resources/whats-new/personalize-ads-with-amazon-dsp-responsive-e-commerce-creatives)

• Yahoo’s native DCO study (US traffic; 173,000 impressions in online bucket testing) reported a 53.5% CVR lift when optimized combinations were used vs. random serving.
  (Source: Dimitrios Kotzias et al., “Dynamic Creative Optimization in Native Advertising,” arXiv preprint 2211.11524, Yahoo, 2022, https://arxiv.org/abs/2211.11524)

A separate marketing field experiment with AI-generated banner ads (global web sample; 173,000 impressions) showed AI imagery can reach up to 50% higher CTR than conventional stock photos in some segments.
(Source: Alexander Junge et al., “Generative AI in Digital Advertising: Field Evidence from Display Campaigns,” arXiv preprint 2411.00168, 2024, https://arxiv.org/abs/2411.00168)

However, the human team still:

• Chooses what to say and who to target.

• Designs the master creative and brand-safe variants.

• Sets constraints for what AI can mix and match.

In practice:

• DCO + AI excels at testing and scaling variants.

• Human creativity remains essential to define the right space for those tests.

Creative Automation Platforms for Display Advertising — Platform Comparison and Features

Below is a simplified comparison of how different tools typically position themselves for display ad production.

Note: Details are representative of common market offerings as of 2026 and based on vendor claims and public documentation. Always check individual vendor sites for current pricing and capabilities.

| Platform Type | Example Focus | Key Features | Export Formats | Governance & Brand Control | DCO Integration |

|--------------|---------------|-------------|----------------|----------------------------|----------------|

| Design tools | Figma, Adobe XD | UX/UI design, prototyping, static layouts | PNG, SVG, design files | Manual styles; not media-governed | None (handoff only) |

| Creative suites | Adobe Creative Cloud | High-fidelity design and motion | Images, video, HTML (via Animate) | Brand libraries; manual enforcement | Indirect via exports |

| Creative automation platforms | Celtra, Smartly.io | Template-based asset scaling, feed-driven creatives | HTML5, video, image | Brand templates, approval workflows | Often integrated or via ad server |

| Production infrastructure (Viewst) | HTML5-native banner studio | AI Smart Resize, AI Image Deflatening, AI Instant Animator, AI Designer; Figma/Adobe import | HTML5, GIF, MP4 | Brandbooks, locked styles, integrated reviews | Designed to output ad-server-ready files; used alongside DCO |

Viewst specifically positions itself as:

• Sitting between design (Figma, Adobe) and media (DSPs/ad servers).

• Owning the high-friction production layer: scaling, animating, approving, and exporting.

• Providing native editable HTML5 output, so ad ops and media teams receive network-ready files with editable layers.

For teams comparing creative automation tools for display advertising:

• Use design tools for ideation and UX.

• Use Viewst or similar production studios for banner scaling and HTML5 production.

• Use DCO platforms for serving and optimizing variants.

Practical Workflow: Humans vs. AI in Banner Production

A realistic division of labor looks like this.

Humans handle:

• Campaign strategy and goals

• Concept, narrative, and messaging hierarchy

• Master creative in Figma/Adobe

• Brand rules, legal constraints, and ethical standards

• Final approval and sign-off

AI and creative automation tools handle:

• Smart resizing into IAB formats (e.g., 300x250, 728x90, 160x600)

• Routine animation (fades, slides, reveals) via presets

• Automated exports (HTML5, GIF, MP4) per platform spec

• Bulk variant generation for:

  • Copy testing

  • Localization

  • Audience segments

Viewst’s infrastructure role

Within this workflow, Viewst acts as:

• Production studio for display ads
  Figma/Adobe master in → HTML5 banner set out.

• Governed automation layer
  Brandbooks, locked components, and single source of truth for master creatives.

• Collaboration hub
  Comments and approvals live inside the banner set, not in screenshots.

This keeps creative judgment human while making production fast, predictable, and scalable.

Actionable Steps for Heads of Creative and Production Leads

If you're responsible for shipping high-volume display campaigns, here’s how to align expectations with reality.

1. Define your “human-only” zones

Document tasks where human review is mandatory:

• Brand positioning and strategic messaging

• Sensitive or identity-related visuals

• Regulatory and legal compliance

• Final sign-off for key campaigns

Align this with IAB guidance on material-based AI disclosure.

2. Map your production pain points

List recurring tasks that cause burnout:

• Resizing and spec checks

• Exporting and asset packaging for each channel

• Manual animation work that doesn’t affect concept

• Versioning for markets, languages, and tests

These are prime candidates for banner ad production automation.

3. Choose infrastructure, not just tools

When reviewing creative automation software:

• Look for HTML5-native output, not only flat images.

• Ensure there is brandbook-level control over fonts, colors, and components.

• Check for integrated review and approvals inside the production environment.

Viewst is one such infrastructure option when:

• Speed-to-market and brand consistency are non-negotiable.

• You run multi-market campaigns with frequent iteration.

• You already use Figma/Adobe and need frictionless handoff into HTML5.

4. Integrate with DCO, don’t replace it

Use AI creative automation to:

• Generate clean, governed variants.

• Export HTML5/GIF/MP4 in ad-server-ready formats.

Then:

• Feed these variants into your DCO or DSP.
  Let those platforms handle delivery and optimization.

5. Establish AI governance and disclosure

Based on IAB’s framework:

• Decide when you disclose AI participation in creative.

• Set rules for data isolation and human review (similar to BrandComms.AI’s standards: human review, client data isolation, output validation).
  (Source: BrandComms.AI, “AI Policy and Responsibility,” 2025, https://www.brandcomms.ai/aipolicyandresponsibility)

This protects both your brand and your teams.

FAQ: AI in Display Ad Creative

1. Can AI fully replace human designers in display ad production?

Current evidence suggests AI cannot reliably replace human designers across the full creative lifecycle.

Studies show AI-supported designs can be more creative and unconventional but not consistently better on brand alignment or usefulness (Kim et al., 2024, arXiv:2411.00168).

In practice, AI is strongest at scaling and variation, while humans remain essential for brand judgment, storytelling, and final approval.

2. What are the best use cases for AI in display advertising?

The most effective uses today are:

• Automating HTML5 banner creation from a master design

• Automated asset resizing into standard display formats

• Bulk variant generation for A/B tests and localizations

• Routine animation and export tasks

These use cases align with Adobe’s findings that creatives use AI heavily for ideation and design, but retain human control for client-facing decisions.

3. How does Viewst differ from general AI design tools?

Viewst is positioned as HTML5-native production infrastructure for display ads, not a general-purpose design or image-generation tool.

Vendor-described differences include:

• AI features (Smart Resize, Image Deflatening, Instant Animator) wired around a master creative.

• Native editable HTML5 output for ad networks.

• Brandbooks and governance enforcing typography, color, and layout rules.

• Integrated review and approval inside the production studio.

4. How should we think about ethical AI in advertising creative tools?

Following IAB and industry guidance:

• Use material-based disclosure when AI significantly influences authenticity, identity, or representation.

• Maintain human review for sensitive content and final approvals.

• Implement brand governance (locked brandbooks, approved component libraries) so AI operates within guardrails.

Tools like Viewst can help operationalize these rules by embedding brand standards directly in the production environment.

5. Do AI-driven banner ads actually perform better?

Performance depends on context, but field studies show promising results.

Yahoo’s native DCO tests reported a 53.5% CVR lift from optimized creative combinations vs. random serving (Kotzias et al., 2022).

A separate study found AI-generated banner images can reach up to 50% higher CTR than traditional stock photography in certain campaigns (Junge et al., 2024).

These results assume strong human-side creative foundations, with AI focusing on optimization and scaling rather than end-to-end concept creation.

The reality is straightforward:

• AI is already a powerful production multiplier.

• Human creatives remain responsible for strategy, brand integrity, and ethical judgment.

• The teams that win treat AI-powered platforms like Viewst as infrastructure for banner production, not as creative replacements.