PhotoShelter Unveils AI Visual Search Capability Eliminating Need for Metadata

Digital asset management platform PhotoShelter has unveiled a new feature called AI Visual Search, giving brand subscribers the ability to search their entire brand library based on visual descriptions and eliminating the need to manually add metadata to images, according to a news release.

Manually searching DAM platforms for the right visual asset can take hours, particularly for larger organizations whose brand library may contain hundreds of thousands of visual assets. PhotoShelter said 40% of the content on its servers had no metadata.

“Even the most advanced content organizations struggle to set up and maintain a comprehensive metadata strategy,” PhotoShelter said. “It would take an estimated 200 days to manually tag every image uploaded to PhotoShelter each month, and metadata-less search eliminates this incredible time-consuming challenge.”

AI Visual Search uses language and image recognition to “intelligently match” search terms to visual qualities, according to the announcement. 

“What was once a manual, difficult and time-consuming task is now a prime use case for leveraging AI to maximize efficiencies and bring better content to market,” said CEO Andrew Fingerman. “And the results are breathtaking – it’s like every conceivable keyword and caption has been applied automatically to our customers’ images.”

PhotoShelter noted that it recently launched a formal Ethics Board to routinely consider the ethics of its AI usage, improve training models, and remove any unconscious bias. “The company will also be aligning closely with the Content Authenticity Initiative to track source images and determine if AI has been used in their creation to monitor and get ahead of disinformation and authorship issues,” PhotoShelter said.

AI Visual Search joins several other AI-based PhotoShelter solutions addressing the challenges of tagging visual assets with metadata: 

ObjectID automatically creates metadata tags for images based on the objects in them, for example “football” or “X-ray.” BrandID, PeopleID, and RosterID tag people, player rosters, and brands such as a NFL team, a high school baseball player, or a coach or other face familiar among the subscriber’s asset library.

Learn more at Brands.PhotoShelter.com.


About the Author

Kristal Kuykendall is editor, 1105 Media Education Group. She can be reached at [email protected].


Featured

  • From the Kuali Days 2025 Conference: A CEO's View of Planning for AI

    How can a company serving higher education navigate the changes AI brings to ed tech? What will customers expect? CT talks with Kuali CEO Joel Dehlin, who shared his company's AI strategies with attendees at Kuali Days 2025 in Anaheim.

  • glowing blue AI sphere connected by fine light lines, positioned next to a red-orange shield with a checkmark

    Cloud Security Alliance Offers Playbook for Red Teaming Agentic AI Systems

    The Cloud Security Alliance has introduced a guide for red teaming Agentic AI systems, targeting the security and testing challenges posed by increasingly autonomous artificial intelligence.

  • Training the Next Generation of Space Cybersecurity Experts

    CT asked Scott Shackelford, Indiana University professor of law and director of the Ostrom Workshop Program on Cybersecurity and Internet Governance, about the possible emergence of space cybersecurity as a separate field that would support changing practices and foster future space cybersecurity leaders.

  • abstract pattern of cybersecurity, ai and cloud imagery

    OpenAI Report Identifies Malicious Use of AI in Cloud-Based Cyber Threats

    A report from OpenAI identifies the misuse of artificial intelligence in cybercrime, social engineering, and influence operations, particularly those targeting or operating through cloud infrastructure. In "Disrupting Malicious Uses of AI: June 2025," the company outlines how threat actors are weaponizing large language models for malicious ends — and how OpenAI is pushing back.