Local-first processing
The audited extension runtime performs supported detection locally, without an AETHER backend, external detection request, analytics, telemetry or prompt history.
Full technology & asset acquisition
AETHER is an early-stage browser security project built to reduce accidental disclosure of selected sensitive-data formats before submission to ChatGPT.
The problem
Emails, phone numbers, selected credential formats and financial identifiers can be pasted into an AI conversation without deliberate intent. AETHER adds a local control immediately before submission.
The audited extension runtime performs supported detection locally, without an AETHER backend, external detection request, analytics, telemetry or prompt history.
A sensitive first attempt is stopped, protected text is written into the editor, and a second explicit user action is required. There is no automatic resend.
Reviewed formats include email, phone, selected API credentials, 89-country IBAN validation and selected payment-card networks. AETHER is not presented as universal DLP.
Verified technical achievements
The acquisition package preserves the RC1 baseline, checkpoint history, test evidence, privacy review and known limitations. Historical results are time-bounded and should be reproduced by an acquirer before a new release.
The canonical 1.0.0 RC1 archive passed integrity review, with all 103 internal file hashes reverified.
Final evidence records 361/361 unit and audit tests, 68/68 compiled E2E tests and 11/11 payment-card-specific E2E tests passed.
A High-severity payment-card false negative initially blocked release, was corrected in a focused change and was followed by regression evidence.
Checkpoint archives, rollback snapshots, audit reports, test suites, privacy review and technical-debt records are preserved for qualified diligence.
Proposed transaction perimeter
Inclusion remains subject to final ownership, licensing and transferability verification.
Proposed assets
Explicitly excluded
Current development status
The transaction materials separate the approved extension baseline from incomplete or unvalidated web components.
Strategic possibilities
These are opportunity hypotheses, not evidence of existing commercial traction.
Independently validate the existing extension and evaluate integration into broader DLP, endpoint-security or secure-workspace products.
Explore the local send-boundary approach within AI safety, governance or enterprise enablement workflows.
Reproduce the frozen release, complete independent assurance and determine the appropriate distribution and commercialization path.
Transaction
Qualified buyers may request staged access to the approved acquisition overview, technical due-diligence report, asset inventory and controlled technical evidence after buyer screening and appropriate confidentiality arrangements.