“I don’t think we can stop the proliferation of deep fake, but there should be some way to digitally watermark or co-stamp authentic videos - so that at least we can tell the difference”
Consumers view digital watermarks as a necessary but imperfect tool for authenticity and copyright protection, with significant frustration directed at visible branding on repurposed content and the ease of removing such marks.
Consumers associate digital watermarks with copyright protection, professional branding, and AI content transparency, but they also object to visible watermarks as aesthetic drawbacks and to invisible watermarks as removable or bypassable. Creators value watermarks for safeguarding photography and signaling professionalism, while platforms and experts promote invisible watermarks as a robust layer for tagging AI-generated content. However, the same consumers actively seek removal tools for cleaner visuals, and some express frustration when platform policies restrict watermark-free saving. The core tension is between protection and visual purity, with trust in invisible watermarks undermined by perceived ease of removal.
Users see invisible digital watermarks as a critical technical solution for verifying AI-generated media and distinguishing real from fake content in an era of deepfakes.
“I don’t think we can stop the proliferation of deep fake, but there should be some way to digitally watermark or co-stamp authentic videos - so that at least we can tell the difference”
“every image made by the generator has a unique digital watermark that identifies it”
“Enter SynthID by Google DeepMind It can produce a digital watermark that's imperceptible to humans and works across Google products to tag AI generated images video, audio, and text”
Creators actively use visible watermarks to assert ownership, deter theft, and maintain professional branding across their visual work.
“Watermarking isn’t just about slapping your name on a photo. It’s about branding, protection, and professionalism... It helps prevent others from stealing or misusing your images without giving credit”
“First, consider watermarking your images or videos... even a subtle watermark can serve as a deterrent and a clear marker of your ownership”
“Plus I got to test out my brand new watermark, which I will be using for all of my art from now on... The choice to implement a new watermark is an important step for digital artists to protect their creations against unauthorized use”
There is a strong consumer desire to remove watermarks, particularly platform logos like TikTok's, to improve engagement and aesthetic quality, leading to widespread use of removal tools.
“But then you download it, and there's that pesky TikTok watermark, bouncing around, ruining the vibe... I notice a huge difference in engagement when I manage to save it without the watermark”
“Removing watermarks can be a common challenge... Canva's Magic Eraser tool is a game-changer for this because it allows ... to efficiently remove unwanted elements, such as watermarks”
“with watermark removal tools, Kaza.ai came as a breath of fresh air... ensures that doesn't happen [ruin an image’s appeal]”
Current products and platform claims respond to consumer concerns by offering both visible and invisible watermarking solutions, with invisible watermarks positioned as harder to remove and cross-platform compatible. For example, Google DeepMind's SynthID is described as producing imperceptible watermarks for AI-generated content across images, video, audio, and text, while TikTok highlights its own invisible watermark as a safeguard. However, these claims do not address the consumer demand for watermark removal tools, which are actively marketed by third-party apps. The market response is split: protection features are promoted, but removal tools remain widely available, indicating a gap between platform intentions and consumer behavior.
Visible watermarks ruin the aesthetic appeal and native feel of content, negatively impacting engagement on platforms like Instagram.
“But then you download it, and there's that pesky TikTok watermark, bouncing around, ruining the vibe... I notice a huge difference in engagement when I manage to save it without the watermark”
“First, consider watermarking your images or videos... even a subtle watermark can serve as a deterrent and a clear marker of your ownership”
Watermarks are easily removed by consumers using AI tools or manual editing, undermining their effectiveness as a deterrent against theft.
“Removing watermarks can be a common challenge... Canva's Magic Eraser tool is a game-changer for this because it allows ... to efficiently remove unwanted elements, such as watermarks”
“with watermark removal tools, Kaza.ai came as a breath of fresh air... ensures that doesn't happen [ruin an image’s appeal]”
Current visible watermarking methods can be cropped out or obscured, failing to provide robust proof of ownership.
“First, consider watermarking your images or videos... even a subtle watermark can serve as a deterrent and a clear marker of your ownership”
The gap lies between consumer expectations for clean, watermark-free content and the protective purpose of watermarks. Consumers want to protect their work but also want to remove watermarks from others' content for aesthetic or workflow reasons, creating a contradiction. Additionally, while invisible watermarks are promoted as robust, consumers and third-party tools demonstrate that removal is possible, undermining trust in AI transparency claims. Platform policies that restrict watermark-free saving further frustrate users, indicating that current offerings do not balance protection with user control. The expectation is for seamless, non-intrusive protection that is also tamper-evident, which is not yet fully realized.
Formulation approaches for digital watermarks focus on invisibility and robustness, as seen in claims about watermarks that are imperceptible to humans and designed to be harder for others to remove. These approaches aim to embed watermarks in a way that survives editing or redistribution, addressing the objection that visible watermarks ruin aesthetics. However, the evidence does not detail specific technical methods, and the persistence of removal tools suggests that current formulations are not fully tamper-proof. The emphasis on 'robust technological watermarks' indicates a response to trust concerns, but without patent or product data, the exact formulation mechanisms remain unclear.
A consumer-led opportunity exists for developing watermarking solutions that are both invisible and resistant to removal, addressing the primary objection of aesthetic disruption while maintaining trust. The demand for removal tools indicates a market for 'clean' content, but also a risk that any watermark can be stripped, undermining authenticity. Researchers should consider that consumers may accept watermarks if they are truly imperceptible and do not interfere with cross-platform sharing. The risk is that overemphasis on protection could drive users toward removal tools, so the opportunity lies in creating watermarks that are tamper-evident without being visually intrusive.