Experts Offer Tips for Securely Deploying Generative AI in the Enterprise

Analysts from Omdia joined AI Business for an interactive webcast this week on Generative AI in the Enterprise – looking at ensuring data privacy and security when utilizing generative AI technologies.

The panel offered tools and frameworks to try when deploying generative AI. Here are their unbiased picks:

--- Missed the live webcast? Sign up to receive it on-demand.

Bradley Shimmin, chief analyst, AI and data analytics

Appen - https://appen.com/

“Focused on fine-tuning. Also, Appen offers Red-Teaming services with domain experts testing for vulnerabilities and stressing established safeguards.”

Robust Intelligence - https://www.robustintelligence.com/

“Focuses on privacy/security through tools like differential privacy and federated learning.”

Dromedary - https://github.com/IBM/Dromedary

“A large language model and large language model development framework designed to help companies create self-aligned models using local data.”

Aporia - https://www.aporia.com/

“Specializes in regulatory requirement compliance across security, privacy and large language model output/outcomes.”

Arthur - https://www.arthur.ai/

“Using a pre-built test suite, customers can weigh the impact changes will have at a token level before moving to deployment; this same mechanism also provides gating for model outputs that breach established thresholds—a literal firewall for large language models.”

Related:Privacy Reigns: Analysts Emphasize Oversight of Generative AI

Curtis Franklin, principal analyst, enterprise security management

Darktrace - https://darktrace.com/

“The Antigena AI uses artificial intelligence to protect applications including generative AI.”

Microsoft Counterfit - https://www.microsoft.com/en-us/security/blog/2021/05/03/ai-security-risk-assessment-using-counterfit/

“Aids in security testing for AI infrastructure. Used by Microsoft red teams.”

Dynatrace AIOps - https://www.dynatrace.com/platform/aiops/

“While this can be used to protect any number of cloud-based applications and services, it is being deployed as a tool for protecting cloud-hosted AI instances.”

Andrew Brosnan, principal analyst, AI applications in life sciences

Federated and swarm learning tools:

Nvidia FLARE - https://developer.nvidia.com/flare

HPE Swarm Learning - https://www.hpe.com/us/en/hpe-swarm-learning.html

Owkin Substra - https://www.owkin.com/substra

Rhino Health - https://www.rhinohealth.com/

IBM and Microsoft also have solutions. 

Ben Wodecki, junior editor, AI Business

Project IDX from Google - https://idx.dev/

“IDX is a new browser-based workspace for developers. You get to try out AI coding tools and you can access existing applications from GitHub. There’s no Python as of yet, but that’s coming – just sign up to the waitlist to gain access.”

Related:Half of Businesses Using AI Report Positive Results

--- Missed the live webcast? Sign up to receive it on-demand.