{"id":33088,"date":"2025-09-09T19:59:35","date_gmt":"2025-09-09T19:59:35","guid":{"rendered":"https:\/\/www.hotelsalepage.com\/feed\/cision-pr-newswire\/desilo-and-cornami-launch-encrypted-ai-model-that-balances-privacy-and-performance\/"},"modified":"2025-09-09T19:59:35","modified_gmt":"2025-09-09T19:59:35","slug":"desilo-and-cornami-launch-encrypted-ai-model-that-balances-privacy-and-performance","status":"publish","type":"post","link":"https:\/\/thaipropertynews.com\/feeds\/?p=33088","title":{"rendered":"DESILO and Cornami Launch Encrypted AI Model That Balances Privacy and Performance"},"content":{"rendered":"

SAN JOSE, Calif.<\/span>, Sept. 10, 2025 \/PRNewswire\/ —\u00a0DESILO, a privacy-enhancing technology (PET) startup, and Cornami, a leader in scalable computing architectures<\/span>, announced the deployment of a fully homomorphic encryption (FHE)-based large language model (LLM) at the AI Infra Summit 2025<\/b>. The solution processes sensitive data while it remains encrypted, delivering both speed and accuracy in real-world use cases.<\/span><\/p>\n

\n

\"At <\/a>
At the AI Infra Summit 2025, Desilo and Cornami Launch Encrypted AI Model That Balances Privacy and Performance.<\/span><\/p>\n<\/div>\n

Addressing the Privacy\u2013Performance Tradeoff<\/b><\/p>\n

Enterprises in healthcare, finance, and other sectors have long faced a tradeoff between strong data protection and AI performance. Traditional encryption methods often slow computation, while faster inference can expose sensitive data. The Cornami\u2013DESILO collaboration seeks to close this gap by enabling encrypted AI inference at speeds close to plaintext processing.<\/p>\n

“For decades, FHE was considered too slow for practical deployment. By accelerating encrypted computation, we are proving that enterprises no longer need to choose between privacy and performance,” said Dr. Craig Gentry<\/span><\/b>, Chief Scientist of Algorithms at Cornami and widely regarded as the father of FHE. Particularly Dr. Gentry highlighted the role of plaintext ciphertext matrix multiplication (PCMM), which affects more than 90% of LLM computation. According to him, PCMM allows matrix operations \u2014 the backbone of LLM processing \u2014 to be executed securely with minimal overhead. Combined with Cornami’s scalable Compute Fabric, it delivers performance that is orders-of-magnitude faster than traditional encrypted approaches.<\/p>\n

“The value of PCMM,” Gentry explained, “is that it makes privacy-preserving AI practical. By enabling matrix multiplication \u2014 the core of LLM computation \u2014 to be performed efficiently and securely, we are closing the performance gap between fully encrypted and plaintext LLM inference, while also strengthening compliance, data sovereignty, and post-quantum security.”<\/p>\n

Applications Across Industries<\/b><\/p>\n

Healthcare illustrates a clear use case. Clinical trials generate large volumes of imaging and patient data that must remain secure, yet timely analysis is essential. With the new solution, researchers can analyze encrypted data without exposing underlying records.<\/p>\n

The same approach applies to finance, government, and cloud services, where compliance with data sovereignty and zero-trust requirements is essential. “Our principle has always been simple: never decrypt,” said Seungmyung Lee<\/b>, CEO of DESILO. “With Cornami, that principle becomes practical. This collaboration lays the foundation for enterprises to safely unlock high-value data, and it aligns with the upcoming launch of our HARVEST\u2122 platform, which will support global healthcare partners.”<\/p>\n

About Cornami<\/b><\/p>\n

Cornami delivers breakthrough performance for scalable computing, enabling advanced\u00a0encryption technologies like FHE to operate at speeds previously unattainable. Its architecture is\u00a0designed to accelerate privacy-preserving AI across industries that demand both security\u00a0and scale.<\/p>\n

About DESILO<\/b><\/p>\n

DESILO is a Korea-based PET startup scaling globally with deployments across Asia<\/span> and\u00a0upcoming pilots in North America<\/span> and Europe<\/span>. The company develops secure collaboration\u00a0solutions using FHE and federated learning, ensuring sensitive data never leaves encrypted\u00a0form. Its flagship Harvest\u2122 <\/b>platform, launching this December, accelerates secure\u00a0multi-party data collaboration in healthcare and beyond.<\/p>","protected":false},"excerpt":{"rendered":"

<\/p>\n

SAN JOSE, Calif.<\/span>, Sept. 10, 2025 \/PRNewswire\/ —\u00a0DESILO, a privacy-enhancing technology (PET) startup, and Cornami, a leader in scalable computing architectures<\/span>, announced the deployment of a fully homomorphic encryption (FHE)-based large language model (LLM) at the AI Infra Summit 2025<\/b>. The solution processes sensitive data while it remains encrypted, delivering both speed and accuracy in real-world use cases.<\/span><\/p>\n

\n

\"At <\/a>
At the AI Infra Summit 2025, Desilo and Cornami Launch Encrypted AI Model That Balances Privacy and Performance.<\/span><\/p>\n<\/div>\n

Addressing the Privacy\u2013Performance Tradeoff<\/b><\/p>\n

Enterprises in healthcare, finance, and other sectors have long faced a tradeoff between strong data protection and AI performance. Traditional encryption methods often slow computation, while faster inference can expose sensitive data. The Cornami\u2013DESILO collaboration seeks to close this gap by enabling encrypted AI inference at speeds close to plaintext processing.<\/p>\n

“For decades, FHE was considered too slow for practical deployment. By accelerating encrypted computation, we are proving that enterprises no longer need to choose between privacy and performance,” said Dr. Craig Gentry<\/span><\/b>, Chief Scientist of Algorithms at Cornami and widely regarded as the father of FHE. Particularly Dr. Gentry highlighted the role of plaintext ciphertext matrix multiplication (PCMM), which affects more than 90% of LLM computation. According to him, PCMM allows matrix operations \u2014 the backbone of LLM processing \u2014 to be executed securely with minimal overhead. Combined with Cornami’s scalable Compute Fabric, it delivers performance that is orders-of-magnitude faster than traditional encrypted approaches.<\/p>\n

“The value of PCMM,” Gentry explained, “is that it makes privacy-preserving AI practical. By enabling matrix multiplication \u2014 the core of LLM computation \u2014 to be performed efficiently and securely, we are closing the performance gap between fully encrypted and plaintext LLM inference, while also strengthening compliance, data sovereignty, and post-quantum security.”<\/p>\n

Applications Across Industries<\/b><\/p>\n

Healthcare illustrates a clear use case. Clinical trials generate large volumes of imaging and patient data that must remain secure, yet timely analysis is essential. With the new solution, researchers can analyze encrypted data without exposing underlying records.<\/p>\n

The same approach applies to finance, government, and cloud services, where compliance with data sovereignty and zero-trust requirements is essential. “Our principle has always been simple: never decrypt,” said Seungmyung Lee<\/b>, CEO of DESILO. “With Cornami, that principle becomes practical. This collaboration lays the foundation for enterprises to safely unlock high-value data, and it aligns with the upcoming launch of our HARVEST\u2122 platform, which will support global healthcare partners.”<\/p>\n

About Cornami<\/b><\/p>\n

Cornami delivers breakthrough performance for scalable computing, enabling advanced\u00a0encryption technologies like FHE to operate at speeds previously unattainable. Its architecture is\u00a0designed to accelerate privacy-preserving AI across industries that demand both security\u00a0and scale.<\/p>\n

About DESILO<\/b><\/p>\n

DESILO is a Korea-based PET startup scaling globally with deployments across Asia<\/span> and\u00a0upcoming pilots in North America<\/span> and Europe<\/span>. The company develops secure collaboration\u00a0solutions using FHE and federated learning, ensuring sensitive data never leaves encrypted\u00a0form. Its flagship Harvest\u2122 <\/b>platform, launching this December, accelerates secure\u00a0multi-party data collaboration in healthcare and beyond.<\/p>\n

<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"rop_custom_images_group":[],"rop_custom_messages_group":[],"rop_publish_now":"initial","rop_publish_now_accounts":[],"rop_publish_now_history":[],"rop_publish_now_status":"pending","footnotes":""},"categories":[5,7],"tags":[],"class_list":["post-33088","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-cision-pr-newswire","category-cision-pr-newswire-en"],"_links":{"self":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/posts\/33088","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=33088"}],"version-history":[{"count":0,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/posts\/33088\/revisions"}],"wp:attachment":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=33088"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=33088"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=33088"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}