Human-Computer Symbiosis Run Amok?

In J. Licklider’s 1960 essay “Man-Computer Symbiosis” he envisions a coupling of the human brain and machine resulting in a partnership in which “no human brain has ever thought”, able to process data much more effectively than any then-known computer. But I wonder would he (or any of us for that matter) desire the truly symbiotic relationship he proposes? In ecology, symbiosis refers to a community relationship between two different species (dissimilar organisms). Though he uses symbiosis to represent a relationship in which both species benefit (what biologists call a +/+ relationship), a more accurate term is mutualism. In this case, he is using “symbiosis” (better stated as mutualism) only in the sense that the thing we created continues to “thrive” and better serve us. Much like saying that our dairy cattle are sharing a symbiotic relationship with us: We built their habitats, we breed them, feed them and raise them to produce a consistent supply of milk. And when they have expired their purpose, we may well serve them up as main course at dinner, replacing them with younger, more productive models. Yes, the cattle do benefit from our care, but there is no genuine give/take in which they are choosing to participate in the community interaction for personal gain. All benefit is measured in terms of their usefulness to us.
And I don’t think anyone would argue we truly desire otherwise. Even Licklider, with his rather romantic musings on “interaction with a computer in the same way that you think with a colleague whose competence supplements your own”, and man and computer being able to “draw pictures and to write notes and equations to each other”, recommends the computer “accept a clearly secondary status in those areas”. As though it’s a good friend, confidante and even advisor, albeit one that is resigned to being less than our equal. Sort of like the family pet but with creative inputs? I can’t help but apply another biological construct: the alpha male. Computers as intellectual counterparts are super as long as they know their place in the social hierarchy!
I recently watched a Naked Science program titled “How to Destroy a Planet” in which scientists engaged in hypothesizing ways that earth could truly be undone. One of the scenarios involved the very real scientific fields of artificial intelligence and “self-replicating” robots. While each of these alone poses some (extremely minor) risk to humans if allowed to run amok; were the two to be combined in one entity under the right conditions, there exists the potential that humans could create a non-living creature capable of learning, adapting, and reproducing itself – at the detriment of all else on the planet. Sort of wacky sci-fi by today’s standards, but then weren’t most of Licklider’s ideas at the time he put them on paper?