When several old PCs are available, evaluate them as a group: keep the healthiest candidate, salvage compatible useful parts, securely erase storage and recycle the machines that are inefficient or unreliable.
Donate electronics only when they are genuinely usable, safe and accepted by the recipient. Verify current device rules before drop-off, remove account locks and personal data, and recycle equipment that would only create a disposal burden.
RAM and SSD upgrades make sense for local AI when the existing CPU and platform already meet the workload requirements; they cannot fix missing instruction support, inadequate GPU capability or a platform with no useful upgrade path.
A cheap computer is only a bargain if it runs the required software, uses a supported operating system, is free of activation or management locks, has healthy enough hardware and comes with acceptable return or warranty terms.
Local AI may not justify new hardware when use is occasional, smaller models cannot do the job, current web tools are essential or a large GPU/RAM/storage purchase would mainly duplicate a cloud service.
A UPS works best when it protects the smallest critical infrastructure first: modem or ONT, router, switch, automation controller and selected server loads sized from actual watts and manufacturer runtime data.
A repair-versus-replace decision becomes clearer when hardware repair, Windows support, Linux migration labor, required software, remaining useful life and replacement cost are written into one worksheet.
Kirksville's household hazardous-waste program is not an electronics-recycling program. Use Missouri DNR manufacturer recovery plans, registered recyclers, verified collection events or current retailer take-back programs instead.
Local-AI memory planning starts with the actual model file and quantization, then adds context and runtime overhead while keeping enough RAM and SSD headroom for the operating system and other applications.
A refurbished computer can be a good low-cost local-AI lab when the exact CPU, instruction support, RAM ceiling, storage, GPU options, cooling and power use match the runtime and model you actually intend to test.
A Windows 11 readiness check should separate hard platform limits from upgradeable items such as RAM or storage, then record whether the machine is healthy enough to keep using.