Robot political speech writer

According to Engadget, The University of Massachusetts at Amherst has developed a computer program that can put together political statements (or short speeches, I guess) based on just a few keywords.  However, the actual research paper says, “Despite the good results it is very unlikely that these methods will be actually used to generate speeches for politicians.”

But AI and robots are increasingly in use.  Could this be a case of famous last words?  In other words, who knows what tools political speech writers use behind the scenes when they have to come up with something in a matter of hours.  After all, political speech writer code can be found on GitHub — note that it uses python.

What does a master speechwriter recommend for writing great speeches?  Mario Cuomo (video below) says it is something that can’t be taught, but he does give 4 suggestions.  I’ve listed them here and related each to the robot political speech writer’s abilities.

  1. Read Everything – Since the robot political speech writer puts together sentences using n-grams from 4,000+ political speeches, that’s quite a bit of targeted reading.
  2. Keep it short – Look at the example speech on Engadget’s site.  Pretty short and concise.
  3. Get a second opinion – Presumably whoever might use a robot political speech writer’s output would either edit the results or give it to someone who could.
  4. Connect with the audience – Clearly this is not something a robot would be good at yet, so it’s really up to the politician to do this.

Topology and quantum computing go together

A new approach to handling massively complex problems combines quantum computing and topology.  It could be a huge benefit to those studying complex transportation problems or social networks.  Now, if only it can help financial experts to figure out how to stabilize the world economy.

Some are even suggesting that human consciousness may function through biological topological quantum computers — possibly microtubules, an important part of cellular processes.  This is not, however, the thinking of many A.I. researchers.