Tag Archives: email

Avogadro Corp Book Review / AI Speculation

Avogadro CorpAvogadro Corp: The Singularity Is Closer Than It Appears, by William Hertling, is the first sci-fi book I’ve read with a semi-plausible AI origin story. That’s because the premise isn’t so simple as “increased computing power -> emergent AI”. It’s a much more well defined formula: ever increasing computing power + powerful language processing + never ending stream of training data + goal oriented behavior + deep integration into internet infrastructure -> AI. The AI in the story is called ELOPe, which stands for Email Language Optimization Program, and its function is essentially to improve the quality of emails. WARNING there will be spoilers below, but only enough to describe ELOPe and speculate about how it might be implemented.

What is ELOPe

The idea behind ELOPe is to provide writing suggestions as a feature of a popular web-based email service. These writing suggestions are designed to improve the outcome of your email, whatever that may be. To take an example from the book, if you’re requesting more compute resources for a project, then ELOPe’s job is to offer writing suggestions that are most likely to get your request approved. By taking into account your own past writings, who you’re sending the email to, and what you’re asking for, it can go as far as completely re-writing the email to achieve the optimal outcome.

Using the existence of ELOPe as a given, the author writes a enjoyable story that is (mostly) technically accurate with plenty of details, without being boring. If you liked Daemon by Daniel Suarez, or you work with any kind of natural language / text-processing technology, you’ll probably enjoy the story. I won’t get into how an email writing suggestion program goes from that to full AI & takes over the world as a benevolent ghost in the wires – for that you need to read the book. What I do want to talk about is how this email optimization system could be implemented.

How ELOPe might work

Let’s start by defining the high-level requirements. ELOPe is an email optimizer, so we have the sender, the receiver, and the email being written as inputs. The output is a re-written email that preserves the “voice” of the sender while using language that will be much more likely to achieve the sender’s desired outcome, given who they’re sending the email to. That means we need the following:

  1. ability to analyze the email to determine what outcome is desired
  2. prior knowledge of how the receiver has responded to other emails with similar outcome topics, in order to know what language produced the best outcomes (and what language produced bad outcomes)
  3. ability to re-write (or generate) an email whose language is consistent with the sender, while also using language optimized to get the best response from the receiver

Topic Analysis

Determining the desired outcome for an email seems to me like a sophisticated combination of topic modeling and deep linguistic parsing. The goal would be to identify the core reason for the email: what is the sender asking for, and what would be an optimal response?

Being able to do this from a single email is probably impossible, but if you have access to thousands, or even millions of email chains, accurate topic modeling is much more do-able. Nearly every email someone sends will have some similarity to past emails sent by other people in similar situations. So you could create feature vectors for every email chain (using deep semantic parsing), then cluster the chains using feature similarity. Now you have topic clusters, and from that you could create training data for thousands of topic classifiers. Once you have the classifiers, you can run those in parallel to determine the most likely topic(s) of a single email.

Obviously it would be very difficult to create accurate clusters, and even harder to do so at scale. Language is very fuzzy, humans are inconsistent, and a huge fraction of email is spam. But the core of the necessary technology exists, and can work very well in limited conditions. The ability to parse emails, extract textual features, and cluster & classify feature vectors are functionality that’s available in at least a few modern programming libraries today (i.e. Python, NLTK & scikit-learn). These are areas of software technology that are getting a lot of attention right now, and all signs indicate that attention will only increase over time, so it’s quite likely that the difficulty level will decrease significantly over the next 10 years. Moving on, let’s assume we can do accurate email topic analysis. The next hurdle is outcome analysis.

Outcome Analysis

Once you can determine topics, now you need to learn about outcomes. Two email chains about acquiring compute resources have the same topic, but one chain ends with someone successfully getting access to more compute resources, while the other ends in failure. How do you differentiate between these? This sounds like next-generation sentiment analysis. You need to go deeper than simple failure vs. success, positive vs. negative, since you want to know which email chains within a given topic produced the best responses, and what language they have in common. In other words, you need a language model that weights successful outcome language much higher than failure outcome language. The only way I can think of doing this with a decent level of accuracy is massive amounts of human verified training data. Technically do-able, but very expensive in terms of time and effort.

What really pushes the bounds of plausibility is that the language model can’t be universal. Everyone has their own likes, dislikes, biases, and preferences. So you need language models that are specific to individuals, or clusters of individuals that respond similarly on the same topic. Since these clusters are topic specific, every individual would belong to many (topic, cluster) pairs. Given N topics and an average of M clusters within each topic, that’s N*M language models that need to be created. And one of the major plot points of the book falls out naturally: ELOPe needs access to huge amounts of high end compute resources.

This is definitely the least do-able aspect of ELOPe, and I’m ignoring all the implicit conceptual knowledge that would be required to know what an optimal outcome is, but let’s move on 🙂

Language Generation

Assuming that we can do topic & outcome analysis, the final step is using language models to generate more persuasive emails. This is perhaps the simplest part of ELOPe, assuming everything else works well. That’s because natural language generation is the kind of technology that works much better with more data, and it already exists in various forms. Google translate is a kind of language generator, chatbots have been around for decades, and spammers use software to spin new articles & text based on existing writings. The differences in this case are that every individual would need their own language generator, and it would have to be parameterized with pluggable language models based on the topic, desired outcome, and receiver. But assuming we have good topic & receiver specific outcome analysis, plus hundreds or thousands of emails from the sender to learn from, then generating new emails, or just new phrases within an email, seems almost trivial compared to what I’ve outlined above.

Final Words

I’m still highly skeptical that strong AI will ever exist. We humans barely understand the mechanisms of own intelligence, so to think that we can create comparable artificial intelligence smells of hubris. But it can be fun to think about, and the point of sci-fi is to tell stories about possible futures, so I have no doubt various forms of AI will play a strong role in sci-fi stories for years to come.

Django IA: Registration-Activation

django-registration is a pluggable Django app that implements a common registration-activation flow. This flow is quite similar to the password reset flow, but slightly simpler with only 3 views:

  1. register
  2. registration_complete
  3. activate

The basic idea is that an anonymous user can create a new account, but cannot login until they activate their account by clicking a link they’ll receive in an activation email. It’s a way to automatically verify that the new user has a valid email address, which is generally an acceptable proxy for proving that they’re human. Here’s an Information Architecture diagram, again using jjg’s visual vocabulary.

Django Registration IA

Here’s a more in-depth walk-thru with our fictional user named Bob:

  1. Bob encounters a section of the site that requires an account, and is redirected to the login page.
  2. But Bob does not have an account, so he goes to the registration page where he fills out a registration form.
  3. After submitting the registration form, Bob is taken to a page telling him that he needs to activate his account by clicking a link in an email that he should be receiving shortly.
  4. Bob checks his email, finds the activation email, and clicks the activation link.
  5. Bob is taken to a page that tells him his account is active, and he can now login.

As with password reset, I think the last step is unnecessary, and Bob should be automatically logged in when his account is activated. But to do that, you’ll have to write your own custom activate view. Luckily, this isn’t very hard. If you take a look at the code for registration.views.activate, the core code is actually quite simple:

from registration.models import RegistrationProfile

def activate(request, activation_key):
    user = RegistrationProfile.objects.activate_user(activation_key.lower())

    if not user:
        # handle invalid activation key
        # do stuff with the user, such as automatically login, then redirect

The rest of the custom activate view is up to you.

Django IA: Auth Password Reset

Django comes with a lot of great built-in functionality. One of the most useful contrib apps is authentication, which (among other things) provides views for login, logout, and password reset. Login & logout are self-explanatory, but resetting a password is, by nature, somewhat complicated. Because it’s a really bad idea to store passwords as plaintext, you can’t just send a user their password when they forget it. Instead, you have to provide a secure mechanism for users to change their password themselves, even if they can’t remember their original password. Lucky for us, Django auth provides this functionality out of the box. All you need to do is create the templates and hook-up the views. The code you need to write to make this happen is pretty simple, but it can be a bit tricky to understand how it all works together. There’s actually 4 separate view functions that together provide a complete password reset mechanism. These view functions are

  1. password_reset
  2. password_reset_done
  3. password_reset_confirm
  4. password_reset_complete

Here’s an Information Architecture diagram showing how these views fit together, using Jesse James Garrett’s Visual Vocabulary. The 2 black dots are starting points, and the circled black dot is an end point.

Django Auth Password Reset IA

Here’s a more in-depth walk-thru of what’s going on, with a fictional user named Bob:

  1. Bob tries to login and fails, probably a couple times. Bob clicks a “Forgot your password?” link, which takes him to the password_reset view.
  2. Bob enters his email address, which is then used to find his User account.
  3. If Bob’s User account is found, a password reset email is sent, and Bob is redirected to the password_reset_done view, which should tell him to check his email.
  4. Bob leaves the site to check his email. He finds the password reset email, and clicks the password reset link.
  5. Bob is taken to the password_reset_confirm view, which first validates that he can reset his password (this is handled with a hashed link token). If the token is valid, Bob is allowed to enter a new password. Once a new password is submitted, Bob is redirected to the password_reset_complete view.
  6. Bob can now login to your site with his new password.

This final step is the one minor issue I have with Django’s auth password reset. The user just changed their password, why do they have to enter it again to login? Why can’t we eliminate step 6 altogether, and automatically log the user in after they reset their password? In fact, you can eliminate step 6 with a bit of hacking on your own authentication backend, but that’s a topic for another post.