The Orioles do have Smith and Hyun Soo Kim penciled in as starters, but there's a problem: They are simply not good against left handed pitching. In his career, Smith has hit .202/.282/.312 with a .268 wOBA and a 60 wRC+ in 558 career plate appearances against lefties. Indeed, the deal for Smith actually precipitated the current problem, because it should have been clear to everyone that Smith was a platoon-only player from the outset.
Kim, in obviously much less time (only 23 PA in 2016), has never had a major league hit against against a left handed pitcher. In Kim's case, there's something of a chicken/egg situation: was Kim unable to hit lefties because he can't hit lefties or did the Orioles not give him enough of a chance to prove his ability to hit lefties? Kim was productive in the Korean Baseball Organization against left handed pitching in his career, so there is certainly reason to think that he could see a significant improvement (not hard to do when you don't get a hit in an entire season!) against lefties in 2017. A platoon situation for both corner outfield spots, then, seems like a distinct possibility.
In certain circumstances, platoons can be very effective. In 2012, the Oakland A's tried this strategy in 2012 to great success, and the Hardball Times estimated that teams that use platoons extensively can squeeze an extra win out of their season. Baseball Reference has an interesting tool that shows the percentage of at-bats per team in which the team has the platoon advantage. The Indians led the way last season with a 70% platoon advantage, while the Orioles finished well below league average with just a 48% platoon advantage (read another way, a 52% platoon disadvantage). Given this, it makes sense for the O's to try and make their platoon disadvantages into a strength by combining Kim and Smith with right handed hitters that can mash lefties.
Unfortunately, there are a couple problems with this set up. First, it's not clear that the guys who could platoon with Kim and Smith actually can mash lefties. Joey Rickard is an obvious platoon candidate after putting up an outstanding .861 OPS in limited time against lefties last season. Rickard, however, is dealing with coming back from injury and dramatic second half offensive decline. Craig Gentry has been a league average hitter against left handers in his career and is a good defensive player, but he has fewer than 100 plate appearances in the last two years total. Of the two Rule 5 picks in camp, only Anthony Santander bats from the right side (he's a switch hitter), but he comes with multiple health and effectiveness questions. Mark Trumbo might be an obvious platoon candidate given his experience playing the outfield last season, but he had an atrocious .608 OPS against lefties in 2016 and a somewhat middling .787 for his career, not to mention his well below average defensive work in the outfield last season.
Second, platoons are potentially a drain on the finite resource of roster spots. One platoon can certainly make sense: if Rickard were to put up even an .800 OPS against lefties this year and he started 1/3 of the games with Smith starting the other 2/3, the Orioles could have an All Star caliber aggregate performance in right field, in addition to allowing Rickard to spell Jones and Kim as well. The problem, however, comes from the fact that if Kim also needs a full-time platoon partner, there are four players for two spots. Add in Caleb Joseph and Ryan Flaherty and that's the entire bench. For a team that is fairly obsessed with roster flexibility despite the odd obsession with keeping a Rule 5 pick every season (another potential barrier here), this leaves Buck Showalter with a pretty restrictive bench. Rickard's minor league option helps, but he also may be the best overall platoon candidate on the 40 man roster.
The way forward, at least to start the season, might be rolling the dice with Trumbo as an occasional platoon partner in right and hoping that Kim can handle most of the plate appearances in left. This would give the team flexibility to start Rickard in the minors and give Gentry and some combination of Trey Mancini and a Rule 5 pick a chance to show what they can do. With the news that Michael Bourn is likely out until the end of April, though, the O's may not want to go into the season with only one above average defensive outfielder. Santander's shoulder injury may allow the O's to place him on the 60 day DL, but that simply kicks the can down the road.
In the end, the O's essentially brought this upon themselves. Unless they trust Kim to take a big step forward against lefties this season, the outfield construction simply doesn't make a lot of sense. It's certainly not out of the realm of possibility that increased reps against lefties would allow Kim to become more comfortable with them, similar to his ascension overall last year, but it's a pretty big gamble to take given Seth Smith's long history of awfulness against lefties. The O's can stand to dramatically improve their platoon splits, and in some ways the pieces are there to do it, but it will require a fairly radical commitment that few teams make.