With　the　advent　of　Sixth-Generation　(6G)　wireless　communication　systems,　the　use　of　higher　frequency　millimeter-wave　(mmWave)　and　terahertz　(THz)　bands　is　expected　to　significantly　enhance　system　capacity　and　coverage.　These　high-frequency　operations　enable　large-scale　antenna　arrays,　offering　advantages　in　signal　processing,　target　localization,　and　detection.　Passive　sensing,　which　leverages　existing　communication　signals　such　as　Channel　State　Information　(CSI)　for　target　detection　and　parameter　estimation,　has　gained　attention　in　6G　research.　This　study　introduces　a　novel　AI-based　architecture,　CsiSenseNet,　designed　to　process　CSI　from　multiple　communication　links　for　high-precision　detection　and　localization　of　passive　targets.　Simulation　results　reveal　that　the　system　achieves　sub-meter　localization　accuracy,　with　detection　accuracy　exceeding　90%　for　human-sized　targets　when　using　two　or　more　communication　links.　Larger　target　sizes　and　increased　link　numbers　contribute　to　improved　detection　performance　and　reduced　localization　error.　Compared　to　traditional　angle-based　methods,　CsiSenseNet　demonstrates　superior　accuracy　and　efficiency,　particularly　in　multi-link　scenarios.　These　findings　highlight　the　potential　of　6G-enabled　passive　sensing　in　industrial　and　warehousing　applications,　enabling　smart　manufacturing,　efficient　resource　management,　and　enhanced　privacy.　Future　work　will　focus　on　optimizing　system　performance　for　diverse　indoor　environments　and　complex　deployment　scenarios.　©　2025　Copyright　held　by　the　owner/author(s).