Skip to main content
Back to AI Newsroom
LEGAL

LegalBusinessTechnology

Healthcare AI Implementation: The Sheba-OpenAI Partnership

Reviewed byPhilippe LipschutzPartner

Scope of this briefing: Note: this update is for general awareness and does not constitute legal advice on any specific implementation. Requires attorney consultation for matter-specific guidance.

On 28/07/2026, Sheba Medical Center (via its ARC innovation arm) announced a strategic partnership with OpenAI, becoming the first hospital outside the U.S. to gain access to OpenAI's dedicated healthcare AI suite, including ChatGPT Health. Sheba's own clinical protocols will be integrated into the models, with early API access for research purposes Calcalist/CTech, July 28, 2026 i24NEWS, July 28, 2026.

Why This Matters for Healthcare Institutions and Their Counsel

Data governance & cross-border transfer: patient-derived clinical data and hospital protocols feeding into a U.S. AI vendor's models raises questions under the Privacy Protection Law and Amendment 13 -- including DPO designation, privacy impact assessments before deployment, and data-minimization in training/fine-tuning inputs.

Human oversight & verification: the reasoning of the Ramat Gan Supreme Court precedent (AAA 63194-08-25, Cohen v. Ramat Gan Municipality) holds that administrative/clinical bodies relying on AI outputs must maintain meaningful human review and disclose AI's role where it affects a person. Note: requires attorney verification as to this precedent's precise scope of application to a specific hospital context.

Vendor contract terms: what to scrutinize before signing -- data use/training rights, liability allocation for erroneous AI-assisted recommendations, audit rights, and termination/data-deletion provisions.

Incident response: lessons from unrelated AI security incidents this year underscore the need for defined incident-disclosure protocols with AI vendors.

How CIDAH Can Help

CIDAH has hands-on experience implementing AI systems internally under defined governance -- matter routing, model discipline, and mandatory disclosure protocols. That experience is the practical basis for advising clients on AI-implementation risk, not just theoretical compliance review.

This document combines the firm's advanced AI system with experienced legal expertise, under the close supervision and approval of senior attorney.