This episode was equal parts effective and frustrating. It's also a bit of an event episode, so SPOILERS from here on out.
On the one hand, it's great to finally get some genuine exploration of Airiam's character. Having been seen in the background since the first episodes to take place on the Discovery, her nature has been raised questions among viewers. Although she's obviously artificial to some extent, just based on her appearance, no one knew if she was an android, and alien, some kind of enhanced human... we just couldn't say. Now, at last, we know it was the last option. Airiam turns out to have a tragic superhero backstory: she and her husband were caught in a shuttle crash on their honeymoon, killing him and seemingly leading to her being rebuilt as a cyborg. Basically, she's Starfleet's science-Robocop.
Hannah Cheesman is brilliant in this episode, now that she finally gets to do something with the character. She makes Airiam one of the most human characters in the show, no mean feat behind the thick mask that she wears in every scene save flashbacks. Her obvious anguish as the virus takes her over and she fights for her free will is powerful, and at the times when she is under the control of the programme, she's really quite frightening (the moment when she pulls Nhan's breathing mask off is brutal). It makes the viewer want to see more of her character – and then they kill her off. Now, it's a very well-written and performed death scene, but it lacks the impact it should have because we've only just gotten to know this character. It's as if the TNG writers had only given Tasha Yar three lines of dialogue before writing her into “Skin of Evil” to kill her off. Or perhaps more comparably, if Data, one of the most fascinating characters of the series, hadn't received any character exploration until halfway through season two. Given that Airiam had already become a fan favourite based solely on speculation, making her into an actual character just so they could have a death with some impact comes across as a very cynical move.
That's a shame, since on its own merits, Airiam's storyline here works very well. It's also entirely possible she'll be back – she uploads her memories to the Discovery computer before she goes, so she'd be one of the easier characters to bring back to life. Her nature as a (partially) artificial intelligence ties into the reveal of the Big Bad for the season. It turns out that Section 31 is not only taking tactical information from Control, it's super-intelligent mainframe, but is under its, well, control. I'm of two minds about this revelation. On the one hand, a malevolent AI is a sci-fi cliché, but this is Star Trek and we're allowed to to enjoy the odd cliché along the ride. Essentially, Control is SkyNet, or the Architect from The Matrix (I said the enhanced probe monster looked like one of the bots from that franchise). It's fighting to achieve true sentience, and to wipe out all organic life in the process. We know that it's capable of this, since we've already seen that future can come to pass, and it's this that the Red Angel has returned to prevent.
On the other hand, it makes Section 31 a bunch of heels. It's rather like the rugpull with Captain Lorca last season. Both seasons have set up an interesting villain that raises serious questions about Starfleet and its ethics in wartime. Cornwell even essentially says that Section 31 are there to do the dirty work when war's underway or on the horizon, just as she questioned what it had done to Lorca last season. In both cases, though, the truth is far less interesting. Lorca wasn't the “real” Lorca, he was a moustache-twirling villain from a comically evil parallel universe. Leland and Section 31 aren't really trying to justify their journey down a dark path, they've just been duped by an evil computer. I don't mind the silly cliché plots, but they're being used at the expense of something genuinely interesting.
Cornwell drops other hints as to what might really be going on. At first it looks like the Vulcan Admiral Petar is masterminding events, and Cornwell reveals that she's a logic extremist. This would tie into the backstory of Spock and Burnham, and put an existential crisis at the heart of Starfleet. It's genuinely interesting that not only could an extremist rise so high in Starfleet, but that it's apparently common knowledge, at least among fellow admirals. But no, Control had her killed offscreen, and there goes that plotline.
This was an entertaining episode that could have been a truly great one, but for all its effective moments, as a whole it's ultimately just unsatisfying.
- The Discovery writers don't know the difference between ultraviolet and infrared. And why would video screens intended for human eyes emit either the ultraviolet or infrared signatures of their subjects in the first place?
- Airiam, Detmer and Tilly play a kadis-kot, a game which was mentioned several times in Voyager (Seven of Nine and Naomi Wildman used to/will one day play it). It's a bit unclear in Voyager whether the game originated in the Alpha or Delta Quadrant, but this episode makes it clear it's an Alpha Quadrant invention.
- It's confirmed that Lt. Nhan, the new security chief, is a Barzan (the race from the TNG episode, “The Price”). This is interesting, as they still don't have their own space travel in the 24th century, although they're in diplomatic contact with warp-capable civilisations.
- A Starfleet AI achieving sentience isn't a first, but it could be tied into Zora, the evolved version of the Discovery computer who starred in the Short Trek “Calypso.”
- Although it could still be Burnham, I'd lay 2/1 odds that the Red Angel turns out to be a resurrected Airiam.