The DARPA Hundred Year Starship program is holding a public symposium in Orlando, FL at the end of next month. The symposium will plan a century-long program to launch manned missions to the stars.

Let’s cut to the chase, shall we? In reality no human being will ever embark upon an interstellar journey. Forget those Star Trek images, those stardrive-powered caves of steel with their space-marines and highly-trained crew. Manned starflight is as likely as a human swimming unaided between Europe and America. It will never happen because it will never need to.

There are more than 30 stars similar to the sun within 50 light years, but all except Alpha Centauri A and B are further than 10 light years away. Suppose our ‘Terrestrial Planet Finder’ telescopes discover a star with a promising planetary system, what are the options for its ‘best candidate planet’?

1. The planet could be a rocky wilderness like Mars, devoid of life.
2. There’s life but it’s microbial in the oceans, like earth in its first billion years.
3. Plants and animals abound, but there’s no sign of intelligence.
4. There’s pre- or low-technology intelligent life, like on Earth before 10,000 years ago.
5. There’s a fully-fledged technological civilization.

I put it to you that one year after discovery, no-one is going to care in the slightest about cases 1-4 unless they are scientists in the relevant disciplines. When was the last time you accessed Google Mars? And if we discovered another unknown tribe in the Amazon jungle would you really read past the headline?

Here is how we would actually put an interstellar R&D program together. First we would take advantage of the fact that photons are already travelling from the exo-planet back to Earth. With a suitable space telescope (far easier to build and more useful than a human-ferrying starship) we could image the planet at a resolution similar to today’s satellite imaging of the Earth. With spectroscopic and image processing this would already tell us quite a lot.

In cases 2-4 where there was evidence of somewhat mundane life, we would send probes. These would be instrumented robotic explorers which could carry out every imaginable mode of reconnaissance, exploration and surveying. We could boost them at ultrahigh acceleration to cut the transit time, perhaps using high-power lasers. Once we had the results back, we could create a virtual reality version of the planet which would be indistinguishable from actually being there. No-one would need to travel tens of light years to get the local experience, which is just as well as there would be no business case.

This leaves the final option: someone interesting already lives there. Now we have a problem of etiquette – probes screaming into their system might be misunderstood. We need to send an Ambassador.

The last thing we should do is send human beings. There is nothing a human could do which a future package of AI-based automation couldn’t do far, far better. The Ambassador ship will be stuffed with multi-spectrum surveillance and analysis equipment and the best expert systems we know how to build. Every possible scenario and approach will be exhaustively gamed in advance and built into the AI cores. After all, there is no possibility of real-time assistance from mission control – the round trip communication delay would be decades.

The Ambassador ship will be unarmed save for a reliable self-destruct facility – with no humans on board, losing it is not a disaster. No doubt we will have stealthy and secure ways to make sure that every transaction it has with the aliens will be transmitted back to earth so we find out exactly what happens.

As before, once we have the data we will build advanced virtual reality simulations which will exactly replicate the experience of being there. No-one will have to go when we can update and upgrade the Ambassador ship by light-speed data-links.

So we’ve covered exploration and negotiation, but what about colonization? This will definitely be on the future agenda given a suitable planet, but sending manned star-ships won’t be the way to do it. First we have to do all the exploration already mentioned and then we’ll send the robots to construct a self-sustaining habitat, ideally to terraform the planet. The final step will be to transmit the genetic codes of the future colonists. There is no point sending protoplasm (atoms) when we can send bits. At the far end, artificial wombs and surrogate robotic parents will bring up the first generation of colonists. After that it will be business as usual.

Further reading: The Hundred Year Starship FAQs.