How many times have you encountered an image online and thought, “Is it real or generated by AI?” Have you felt ensnared in a reality where AI-generated and human-created content intertwine? Is it still necessary to differentiate between the two?
Artificial intelligence has opened a realm of creative opportunities but also introduced new challenges, altering how we view online content. With the surge of AI-generated images, music, and videos on social media, not to mention deepfakes and bots deceiving users, AI now permeates a significant portion of the internet.
According to research by Graphite, the volume of AI-generated content exceeded that of human-created content in late 2024, primarily fueled by the introduction of ChatGPT in 2022. Additionally, another study indicates that more than 74.2% of the analyzed pages contained AI-generated content as of April 2025.
As AI-generated content becomes increasingly advanced and nearly indistinguishable from human-created work, society confronts a crucial question: How accurately can users recognize what’s authentic as we move into 2026?
AI content fatigue sets in: Demand for human-created content increases
After several years of enthusiasm surrounding AI’s “magic,” online users have begun to experience AI content fatigue, a collective weariness in response to the relentless speed of AI advancements.
According to a Pew Research Center survey, a median of 34% of adults globally expressed more concern than excitement regarding the rising use of AI in a spring 2025 survey, while 42% felt equally concerned and excited.
“AI content fatigue has been mentioned in various studies as the novelty of AI-generated content fades, and its current form often feels predictable and abundant,” Adrian Ott, chief AI officer at EY Switzerland, shared with Cointelegraph.
“In a way, AI content can be likened to processed food,” he stated, drawing parallels between the evolution of both phenomena.
“When it first emerged, it overwhelmed the market. However, over time, people began to revert to local, high-quality food where they understand the source,” Ott added:
“It might evolve similarly with content. There’s an argument to be made that people prefer to know who is behind the ideas they consume, and a painting is evaluated not just on its quality, but also on the narrative of the artist.”
Ott proposed that labels such as “human-crafted” may arise as trust signals in online content, akin to “organic” in food.
Managing AI content: Approaches to certifying legitimate content
Though many contend that most individuals can identify AI text or images effortlessly, the challenge of detecting AI-generated content is more intricate.
A September Pew Research study found that at least 76% of Americans believe it’s critical to identify AI content, with only 47% feeling confident in their ability to do so accurately.
“While some individuals may fall prey to fake images, videos, or news, others might deny the validity of anything that doesn’t align with their perspective or dismiss real footage as ‘AI-generated,’” EY’s Ott highlighted, pointing out the complexities of managing AI content online.
According to Ott, global regulators are leaning toward labeling AI content, but “there will always be methods to circumvent that.” Instead, he recommended a reverse approach where authentic content is certified as soon as it is created, ensuring authenticity can be traced back to a real event rather than attempting to expose fakes post-facto.
Blockchain’s role in establishing “proof of origin”
“As synthetic media becomes increasingly challenging to differentiate from genuine footage, relying on post-event authentication is no longer effective,” said Jason Crawforth, founder and CEO of Swear, a startup focused on video authentication software.
“Protection will derive from systems that embed trust into content from the outset,” Crawforth noted, emphasizing Swear’s core concept, which ensures that digital media remains reliable from the moment it’s created via blockchain technology.
Swear’s authentication software utilizes a blockchain-based fingerprinting mechanism, linking each piece of content to a blockchain ledger to provide proof of origin—a verifiable “digital DNA” that remains intact without alteration.
“Any alteration, no matter how subtle, can be detected by comparing the content to its blockchain-verified original on the Swear platform,” Crawforth stated, adding:
“Without built-in authenticity, all media, past and present, risks scrutiny […] Swear doesn’t ask, ‘Is this fake?’, it demonstrates ‘This is real.’ That transition makes our solution proactive and future-proof in the quest to safeguard the truth.”
Thus far, Swear’s technology has been adopted by digital creators and enterprise partners, primarily targeting visual and audio media across video-capturing devices, including body cams and drones.
“Although social media integration is a long-term vision, our immediate focus is on the security and surveillance sector, where video integrity is essential,” Crawforth mentioned.
2026 outlook: Platform responsibilities and critical moments
As we move into 2026, online users are increasingly worried about the rising prevalence of AI-generated content and their ability to differentiate between artificial and human-created media.
While AI specialists stress the necessity of clearly identifying “real” content versus AI-generated media, it remains uncertain how quickly online platforms will acknowledge the imperative to prioritize trustworthy, human-made content as AI continues its pervasive expansion across the internet.
“Ultimately, it’s the platform providers’ responsibility to equip users with tools to filter out AI content and highlight high-quality material. If they fail to do so, users will migrate elsewhere,” Ott remarked. “Currently, individuals have limited capacity to eliminate AI-generated content from their feeds — that authority largely lies with the platforms.”
As the need for tools that discern human-made media intensifies, it’s imperative to recognize that the fundamental issue often isn’t the AI content itself, but the motives behind its creation. Deepfakes and misinformation are not entirely new, although AI has significantly escalated their scale and speed.
Related: Texas grid is heating up again, this time from AI, not Bitcoin miners
With only a limited number of startups concentrating on authentic content identification in 2025, the situation hasn’t evolved to a point where platforms, governments, or users are taking immediate, unified action.
According to Crawforth from Swear, humanity has yet to hit the critical juncture where manipulated media inflicts visible, undeniable harm:
“Whether in legal scenarios, investigations, corporate oversight, journalism, or public safety. Delaying action would be unwise; the groundwork for authenticity must be established now.”
