Introducing the Authenticity Market Map

Jeffrey McGregor
October 2024

I have been eager to write this post for nearly a decade. When Truepic was founded in 2015, it became increasingly clear that digital photos were becoming unreliable, and that content authenticity would become critical. At the time, concerns mostly centered around the ease of simple image manipulation and the risk of miscontextualizing content, which posed challenges for many different industries. We did not foresee how rapid the acceleration of AI technology would be, or how existential this threat would ultimately become.

Generative AI has posed a completely unique challenge for trust online. As it stands today, anyone with an internet connection has access to the highest quality LLMs, capable of generating photorealistic images in seconds. Those images can then be distributed widely across social media, 1:1 through messaging apps, or to any company requesting photo documentation. In a matter of months, the internet has become flooded with synthetic content that is, each day, becoming more indistinguishable from authentic media. 

By 2024, more synthetic images were created in a single year than had been produced in the first 150 years of traditional photography. Applications like ChatGPT set records, reaching 100 million users in less than two months. The quality of synthetic imagery has advanced so quickly that we have crossed the uncanny valley, making it nearly impossible to distinguish between human-made and AI-generated content. This transformation has been fueled by substantial capital investment, with Goldman Sachs forecasting over $200B of global investment by 2025. We are just at the start.

Understanding the origin and history of digital media is crucial not only for maintaining a healthy informational ecosystem but also for preserving trust in our digital interactions. We now face the onset of a zero-trust environment for not only photos, but also audio and video—mediums that have historically been vital for how we consume information and shape our worldviews. These trends will be difficult if not impossible to reverse, as parents battle sophisticated deepfake audio fraud and deepfake’s of CFOs lead to multi-million dollar losses.

This evolving digital landscape has given rise to a new category of companies, technologies, and community efforts focused on restoring transparency and preserving the integrity of the information environment we depend on. This category is composed by an exciting mix of innovative startups deploying cutting-edge approaches to authenticity, alongside established enterprises that bring scale and visibility to consumers. We are proud to collaborate with many of the talented individuals leading these efforts and are continually inspired by the remarkable work happening across the industry.

Why we need the authenticity category

People are increasingly overwhelmed by information and more skeptical of the content they encounter online. Consumers are now seeking greater authenticity in their online experiences, wanting assurance that what they see—and base their decisions on—is both accurate and genuine. A Deloitte report reveals that over two-thirds of consumers prioritize authenticity, with millennials—who represent 40% of the market—perceiving authentic brands as moral and accountable. This growing demand for genuine and transparent information is reshaping the marketplace, and as reliance on digital systems increases, this trend will only become more pronounced.

Businesses are recognizing that authenticity is not only a consumer demand but also essential for security. In today’s digital economy, companies cannot remain competitive without ensuring the authenticity of the content that informs their operations. The financial sector, in particular, has become a prime target for deceptive media, which can exploit even minor vulnerabilities with severe consequences. According to The Wall Street Journal, financial institutions are bracing for a future in which synthetic content is routinely used for fraud, resulting in both financial and reputational damage. A recent study of over 1,000 finance professionals found that 85% view synthetic media scams as an “existential” threat to their organization's security.

One of the most concerning areas of visual deception is not with political or even enterprise threat vectors, but rather the local weaponization of it. There is a growing trend of targeting ordinary people, especially young women, whose public images are being sexualized through AI systems. The scale of the problem is concerning, a recent study of over 16,000 respondents across 10 countries found that 2.2% had experienced deepfake pornography victimization and 1.8% admitted to perpetrating it. In South Korea alone, over 500 cases of such attacks have been documented. Moreover, bad actors are broadening their scope, directing malicious efforts toward public servants and vulnerable groups for ideological or nefarious purposes. These images are often spread within direct messaging applications or on anonymous platforms like 4chan, making them very difficult to address in real time. While authenticity measures cannot halt these attacks, they underscore the peril of a global ecosystem where the line between real and synthetic is dangerously blurred, and the absence of clear standards threatens the integrity of our shared digital landscape.

The Authenticity Category

As a result of this significant need, an equally exciting category of companies, solutions, technologies, and communities have emerged—the Authenticity Category—which directly responds to these growing market demands. 

‍

Methodology

Broadly, we see both digital and physical authenticity playing crucial roles in the ecosystem, with solutions often overlapping to solve end-user needs. Within each of these areas, we have defined subcategories that largely align to the technological approaches being applied, all aiming to solve the same core question of… “is this media, tangible, or animate object authentic and/or trustworthy.”

This is not intended to be an exhaustive list of every company in the space, but rather a snapshot of the market to provide insight into the various technologies and approaches that businesses and governments are using to authenticate information today. Our initial goal is to highlight some of the most widely available approaches in the current market. If we’ve missed a company or community, or if you have feedback, please reach out to us at marketmap@truepic.com. We plan to update our methodology, as well as the companies and initiatives included, as the category continues to evolve and mature.

‍

The categories 

‍

Digital Content 

This category is divided into the two core technological approaches to media authenticity, (a) proactive approaches and (b) detection based models. These two approaches are often conflated, but operate very differently. 

• Proactive technologies are applied to digital content to help relay information about its origin and history throughout its lifecycle. These markers are used as indicators downstream to help ascertain the context that surrounds media, including its authenticity.  The three most common approaches are: 
  • Provenance (secure metadata): Best exemplified by the C2PA’s Content Credentials which can be implemented through various vendors. Its critical benefits are interoperability and tamper-evidence. 
  • Watermarking: This involves technologies that add visible or invisible markers to content to trace media downstream or authenticate origin. 
  • Fingerprinting: Unique digital identifiers that are distributed or stored, which help to track, match content, or identify changes over time. While fingerprinting is a valuable technology, it is not pictured as most fingerprinting implementations are proprietary and not regularly offered by 3rd party vendors.
• Detection is also divided into synthetic media detection and Open Source Intelligence Tools (OSINT). 
  • Synthetic media detection: This subcategory focuses on vendors that can ingest and identify AI generated content via running forensics analysis and/or machine learning to identify signals and anomalies that indicate AI-creation. These tools produce probabilistic outcomes to help end-users make a judgment on the likelihood of authenticity.
  • OSINT:  Short for open source intelligence, this category includes various tools which can be used to assess the authenticity of digital imagery. These tools are often associated with the verification of “cheapfakes” or non-AI manipulations to digital content. 

• Standards and Specifications
  • We introduce a variety of bodies, initiatives and communities that are vital to driving authenticity in the digital world. At the forefront is the C2PA (Coalition for Content Provenance and Authenticity), which has been instrumental in creating open standards for digital content provenance, facilitating the verification of media origins and its history—essential tools in the fight against misinformation. Notably, the C2PA's framework is being considered in the ISO Task Force ISO TC 171/SC 2, with the aim of being a global standard in 2025. Other entities like IEC and ERC-7053 focus on standardizing technical infrastructure, while JPEG Trust and ONVIF offer sector-specific protocols for media and security. 
• Communities and Industry Groups
  • Many incredible communities have formed, composed of companies, academia, individuals, think tanks, and more that have strong shared values around the need for trust and transparency in media. Communities such as the Content Authenticity Initiative and Partnership on AI complement the technical efforts happening within the standards and specifications groups by creating the awareness, educational resources, and open source tools to help implement these important technologies.

‍

Physical authentication 

This section focuses on the digital authentication of animate or tangible objects. We often find that physical and digital authenticity is very tightly tied, with vendors from both segments working in unison to help ascertain authenticity through an entirely digital workflow. For example an IDV provider may utilize a digital content authenticity vendor to confirm the authenticity of media prior to running identity verification technologies on the information contained within that media. 

This segment is divided into two main categories: Proof of Human and Physical Goods Authentication.

• Proof of Human: We see authenticating and ensuring human use or testament as an established but ever growing keystone of the authenticity market. We highlight three sub-categories in this section:
  • Identity Verification (IDV) / Know Your Customer (KYC): Vendors that ensure a person using a digital service is who they claim to be by verifying personal information and documents.
  • Biometrics: Vendors that rely on unique physical or behavioral traits, such as fingerprints or facial recognition, to confirm a person's identity. Biometrics are widely used in hardware, especially for smartphone security. They are also often a component of other broader processes (such as IDV/KYC).
  • Captcha: Distinguishes between humans and bots by presenting tasks that are easy for humans but difficult for automated systems. This well-established tool is commonly used in email sign-ups and digital transactions.

‍

• Physical Goods Authentication:  Another highly established sector, many of these technologies are ubiquitous across the globe. Nonetheless, vendors here play a foundational role in ensuring authenticity in the physical assets we buy, ship, send and relay on daily. 
  • QR Codes: A machine-readable code that stores information about a product or service, widely used for tracking, marketing, and verifying the authenticity of physical goods.
  • Blockchain: A decentralized and secure digital ledger that records transactions, ensuring the traceability and authenticity of assets, often being integrated into a variety of verticals in unique ways, 
  • Barcodes: Machine-readable representation of data using a pattern of lines or squares, barcodes store product information that can be quickly scanned to identify, track, and manage items.
  • RFID: A technology that uses radio waves to wirelessly identify and track tags attached to objects, commonly used for inventory management, asset tracking, and counterfeit prevention.
  • Virtual Inspections: A remote inspection process using digital tools, such as photos, videos, or synchronous video, to verify the condition and authenticity of goods without needing physical presence.

‍

Reflections

As we reflect on the work behind mapping this category, there are a few key takeaways worth sharing:

• The authenticity of physical goods, including identity, has been a challenge for much longer than that of digital content. This market segment is well-established and has the market capitalization to validate the commercial value of solving the critical question: “Is it real?”
• Many established solutions for verifying the authenticity of physical goods are now facing a new challenge: maintaining the relevance of their workflows in a world where digital media is increasingly untrusted. We are witnessing a promising convergence between physical and digital authenticity, providing significant value to industries that depend on automated verification.
• Digital content authenticity has surged over the past 24 months, driven largely by advancements in text-to-image generative AI. What was once a category dominated by tools and technologies has evolved into a vibrant space, with vendors now delivering impactful solutions to customers.
• Standards are crucial, and the market is at an ideal juncture where community-driven initiatives and standards organizations can align efforts across companies. This will ultimately lead to interoperable solutions that benefit customers.

‍

This is only the beginning. As we navigate the early stages of AI, we anticipate significant growth across the entire field, as diverse groups—from school teachers grading assignments to intelligence communities analyzing media—grapple with the fundamental challenges of authenticity. While no single solution will address every concern, a coalition of companies is driving forward the mission of authenticity in the AI era. One thing is certain: we remain optimistic that authenticity will prevail on the internet, as the well-being of future generations depends on it.