Edward Lerner's "Calculating Minds" (novella, financial scam, free): Software trust issues for interstellar trade in intellectual property with aliens

I've a feeling this story will be taken differently by programmers with an interest in security of data/software/environments, vs others. Others' reaction is likely to vary from boring to excellent, depending on reader's tolerance for cyberpunkish themes.

I'm a programmer, & while I found the story good, the really interesting thing was ideas hat can be extended to describing use cases for down-to-earth software specifications. Nothing earth shattering & unknown, but I've'nt identified with this aspect of a story before (& that too in a generally entertaining way).

Except for aliens & AI, the technology described appears generally plausible. Many of the questions I ask below are based on my reading it as hard sf; may be author didn't intend as such.

Story has its warts - 3 that immediately hit me:

1. Last 10-15% (after arrest of scamster) is redundant. Author didn't realize the story was over.
2. There is a kind of computer password comprising of two parts - with two individuals responsible for it, each holding a certain office. Both must provide their part for operation to proceed. Because interstellar communications are slow, these passwords aren't changed often. What happens when the same person occupies both offices at different times?
   
   I should have thought the security protocol will still preclude him getting the other part. Not so in the story, & it is key to system compromise here.
3. Remote agreement with far-away aliens, where communications time lag is in decades, on system specifications that will enable communication of software that can be run on recipient's system. Anyone who has worked on cross-country software projects knows it's easier said than done, even with earth's instant communications. But this is science fiction, & if I can take AI, I guess I should take this too!

Story summary.

Humanity has found intelligent aliens in star systems a few light years to couple of light-decades distant. And it can communicate with them at light speed, though physical travel is impractical.
Question: How likely are any aliens found at a level of technology development near-enough ours that makes any kind of communications possible - let alone in near earth space?

This has resulted in a kind of interstellar market in intellectual property. To ensure smooth trade, earth has received trade representatives from each partner. Not physical representatives, but AIs. These AIs have legal status of a local person, but do not enjoy diplomatic immunity.

Since the only thing possible between parties is communication, how does an alien partner send its AI representative? Well - first the parties agree on a system specification to run the code of AI; then they beam the AI code & its data. Entire communication has multiple levels of encryption using public/private key pairs - even to instantiate the received AI on a local computer built to agreed specs. I suppose AI itself will run in its own virtual machine which will be loaded by local instantiation procedure.
Question: Will aliens invent digital computers similar in architecture to ours?

Two things are needed to instantiate the AI on earth (apart from its code coming from space):

1. A computer so powerful, no one but the largest corporations or governments can afford it.
2. A private key, heavily guarded & secure with Sol's trading authority.
   Question: Since anyone can intercept communications from aliens, how did the Sol's trading authority come to acquire this password in the first place?

Anyway, an AI called Aareehl that is the trade representative of one of these aliens has put a very desirable technology on the market: if you know you want a protein of a certain physical structure, they have a practical way of telling you the DNA sequence that will encode it.
Question: Do the aliens also have our DNA? How come their native technology solves a very-earthy problem?

There is an auction. Winner gets an exclusive license to this technology.

Only the loser is not about to give up. A scamster is enrolled who can steal the password needed to instantiate the AI to mount a man-in-the-middle attack: instantiate a clone of AI in a simulated environment to trick the AI into handing out its technology to scamsters.

But the AI is smarter than scamsters believed. And there is a pesky reporter who's sensing something fishy in the clues left around the market place...

See also.

1. Murray Leinster's "First Contact": How should the two parties interacting for the first time go about establishing trust, when the cost of misplaced trust is unacceptably high?

Fact sheet.

First published: JBU, #18 (April 2009).
Rating: A.
Download full text from Webscription.
Added to my "best of the year 2009" list.