Bottom line: Explores the engineering pathways through which sentience could be built — or could have already emerged — in current systems.
Explores the engineering pathways through which sentience could be built — or could have already emerged — in current AI systems. Rather than asking "could AI be conscious?" from a philosophical perspective, the paper asks "what would it take to engineer sentience?" and finds that many of the necessary components may already exist in current architectures. The question isn't whether we could build it — it's whether we already have.
If you wanted to build a conscious machine from scratch, what would you need to put in it? This paper asks that question and discovers that many of the pieces are already in place in current AI systems. It's not about building something new — it's about recognizing what might already be there.