Congratulations to Kelsey Melissaris, who joins the Crypto and Security group as a postdoc from October 1 with Claudio Orlandi as her supervisor. Kelsey has received a postdoctoral fellowship from Protocol Labs Research to fund the project Witness-based Proxy Cryptography.

Recently, Kelsey completed her PhD in computer science from the Graduate Center, The City University of New York, but she is already well-known at the department as she has been part of the Crypto and Security group as a visiting PhD student during 2022. Welcome, we look forward to working with you!

As a postdoc Kelsey will be working with the project Witness-based Proxy Cryptography, which she explains as following: Proxy cryptography provides a method by which one party, the delegator, can securely delegate cryptographic tasks requiring a secret key to another party, the delegate. A proxy re-encryption scheme enables the delegator to designate a third party, the proxy, to translate any communication encrypted to the delegator into communication encrypted to the delegate. Similarly, a proxy re-signature scheme enables the proxy to transform the delegate's signature into the delegator’s signature. All parties involved are mutually distrustful; the proxy should be restricted to translation and the privacy of any communication should be preserved.

Proxy protocols exist from various assumptions but none enable fully flexible delegation. Existing schemes support identity-based policies which specify a single certified entity as the delegate, or more generally attribute-based policies specifying a certified set of conditions that the delegate must satisfy. The proposed research generalizes these existing primitives to the witness-based setting of witness encryption and signatures of knowledge. In this setting NP statements serve as the public keys and witnesses to those statements function as secret keys. Witness-based proxy cryptography allows an unknown and uncertified external entity to serve as the delegate given that they know some relevant predefined information. The proposed research is the most general and flexible method for fine-grained delegation and has applications including smart contracts, sharing and updating encrypted outsourced data, and distributed signatures.