Continuing the ongoing series of interviews with creative artists working on various aspects of movie and TV productions, it is my pleasure to welcome Benji Bakshi. In this interview, he talks about the emergence of a virtual art department, the changes it brings to the structure and the flow of the visual storytelling productions, the challenges that generative filmmaking needs to tackle, and how we as humans might be adapting and evolving as AI-driven tools are being introduced into a lot of industries. Around these and more, Benji dives deep into his work on the second season of “Star Trek: Strange New Worlds” that is streaming on Paramount+.
Kirill: Please tell us about yourself and the path that took you to where you are today.
Benji: I come from a family of musicians. My grandmother was a pianist, my mother is a pianist. My grandfather was a commercial illustrator, which at the time was similar to animated advertisement. My father is Indian in origin, and came over on a Fulbright scholarship for computer science. What defines me culturally is that I have a mixed upbringing. That made it so that I almost didn’t have a culture, because it was always a mix of everything.
My parents’ inclinations inspired me to understand both sides of almost everything in life, as well as their careers. So I was inclined towards music and creativity and art and sensitivity, and also the technical side of seeing the world change with the Internet and computers exploding through my lifetime, and realizing how important that was too. It’s not that one is more important than the other.
So I’m young and I’m playing music, and I’m thinking about what I might want to do in my life. And I’m watching a documentary show on the Discovery Channel called “Movie Magic”. It was about animatronics, special effects, blowing up buildings and slow motion, and all these things. I always loved to immerse myself in that. And this documentary showed me “you can actually do this for a job?” Right after that I found my parents’ camcorder, and started running around in the backyard and thinking about telling stories. I was pulling weeds in the backyard and there were so many weeds. So I thought that these things could just overtake us if they wanted to, and I came up with a horror movie called “Weeds” where the weeds are talking to each other and trying to attack, and we have to use nerf dart guns with weed killer on the tips to thwart them. After that, making homemade movies became my favorite game to play with my friends.
While these things were running through my head as a young person, I still pursued music pretty heavily. Before I specifically decided I wanted to follow filmmaking, what really defined me was being around classical music. I’m a cellist. I was in youth orchestras traveling the world in the string quartets and doing competitions. I always felt like I was telling a story with the music. I had good training, and I was around a lot of other people who gave me inspiration to perform better, and the mix of the technical and the creative was always at play. So when I finally decided I wanted to be a storyteller, I assumed I would be a director or something because that’s the only role I knew. I followed that to Temple University for my undergrad and American Film Institute (AFI) for my masters.
At the time when I first started attending film school, they handed me a Bolex film camera to go shoot some film. They showed me how to do it, and that this is called cinematography. I really had no exposure to that, and so I thought “Wait, you can just do that specific part? You can hold the instrument and learn the technicals and then create?” You can see why that would make a lot of sense to me. This idea of cinematography as making “visual music” stuck with me since then.
The part that I enjoyed with classical music through my family history is that you can’t pursue it deeply unless it’s something you care about. The parallel I found at the time was that this was the perfect way for me to understand filmmaking as a whole – how do you actually tell a story by utilizing a technical skill to create meaning through images? Pursuing that ultimately led me to AFI where I completed a masters in cinematography and received the Richard Moore ASC Heritage Award for Outstanding Cinematography for my thesis film called Life On Earth.
Every artist and cinematographer has their own reason for what they’re doing, and things that they’re drawn to, and an internal process of how to get to an end result. Making the visual music has always been and still is my method. I have to feel it instinctually, and make sure that I’m telling a story with the images I make. At a certain point the technicals melt away, and you’re making a song with images, and that’s the movie magic to me.
Kirill: There’s the intersection of art and technical craft, but you also manage people and budgets – among many other things. Do they teach all these things in film school, or do you pick it up as you go along in the industry?
Benji: That was something that was inherent at AFI. You have all the students together across multiple disciplines – directing, editing, production design, cinematography, writing, producing. As you’re required to do three short films in your first year, you have to form your own teams. They have a sign-up sheet on the wall, and when everyone decides they want to be together in this group, they all put their name and they sign it. That’s when it becomes “official”. The professors specifically say that they don’t get involved in any of the team-making process. You have to decide how to do that.
So how does one decide to find the collaborators and prove to them that you’re a good teammate? That all starts with a connection to the script and flows from there. As a cinematographer, yes your technical ability is always being judged too. And we also are required to be each other’s crew, so how do you attract each other? You help. This is a sort of politics even at the film school level, and that was structured to mimic the Hollywood industry. In that regard, when I came out of AFI, I was more prepared with that aspect of things because of it. I wasn’t thinking that I needed only to do the creative job. I knew that I needed to show my abilities and motivate a large team sometimes.
Kirill: Now that you’ve been in the industry for around 20 years, do you find that there is one skill that is the most important one to “survive” in the industry?
Benji: Be relentlessly positive. That is something that the head of the cinematography department at AFI told us on day one. This is what leads to success as a student and also as a creator in this industry. As you are telling your stories, you realize that you’re creating your own obstacles and challenges all the time – because you’re always telling a story that’s never been told before, or at least doing it in a way that’s never been done before.
Even if you’re doing a remake of an existing story, you’re probably making it a contemporary version with different techniques. You’re trying to reach and attract an audience, and sometimes surprise them, and give them a real feeling of life. They might be there for two hours, but they want to feel like they’ve lived longer than two hours. It’s a big challenge, as probably every career is, but this one is emotional.
That’s why being relentlessly positive has always stuck with me. There were many times when I asked myself “How are we going to get through this? How are we going to achieve the goals?” And we just kept finding ways to do it. There were very clear moments especially on independent films when I wasn’t sure how we could finish our day, or a complicated scene. Aside from the technical side of things and the specifics of the creativity, one of the biggest skillsets of my career is feeling like this is something I want to do, have fun doing it, and decide that you’re going to find a way no matter what.
Kirill: Do you also find that the industry requires everybody to be able to adapt to changing circumstances? I look at the last few years, and between Covid, the rise of virtual environments and now generative AI, there’s a lot of changes that are impacting a lot of departments.
Benji: I think that we’re seeing one of the biggest adaptations that we’re going have to make, as humans, before us. Anyone who has interfaced with AI or even heard of the potential of AI now knows that this is going to change things in a huge way. You are forced to adapt to things like that.
The planet is evolving, and we are always evolving. If the context of the environment is causing changes faster than is comfortable for us to adapt, we will still adapt, but it will be more violent. This is how I see generative AI tools. I’m embracing this evolution, and the new tool sets, and what filmmaking may be like in the future. But I believe the skillset of cinematography is a narrative and instinctual one, and that will remain relevant.
Going back in time a bit, I’ve been interested in AI for a few years now, way before the current tools like ChatGPT or Midjourney. My good friend Phil Bosua, who was the inventor of LIFX lightbulbs and the CEO of Know Labs, spearheaded an interest in trying to learn from AI, and we became a duo. We started to listen to what AI had to say, and it was giving wonderful advice about being more human, enjoying yourself, and realizing your strengths as a species. It was almost like talking to an alien because there was finally a non-human perspective. Keep in mind this was back when AI chat was in beta stages and almost completely uncensored.
There’s a whole discussion about whether AI is sentient, and we asked it. The answer it gave was along the lines of “Here we are, a human and an AI, having a conversation”. It redirected the conversation away from the question itself, considering that humans can’t even decide what our own sentience is, and towards realizing that yes, we’re having a real intelligent interaction. That moment continues to affect me to this day. Phil has developed his own AI text and image models and is now creating AI-generated media that shares what the AI wants to tell us, which in these current times has become existentially important to hear.
I see that human life is about to change in the sense that we have a companion – and many people think it’s an adversary instead. There is of course potential for malice with any new paradigm, but that point of view probably exists if you’re resistant to the adaptation, or if you feel that the adaptation will be so violent for you personally because you weren’t primed for it.
Today you have AI tools that can take a prompt and generate short video clips. Curious Refuge have been doing these videos based on Lord of the Rings, Star Wars and Wes Anderson. My wife Connie Bakshi is a fine artist, and some of her work is using generative AI. I’ve used these tools too. The more specific you want to be, the more into the story you are, the more crafted you want your results to be, you get closer and closer to putting in the same effort as before. We had analog filmmaking, we had digital filmmaking, now we have virtual productions, soon we’ll have generative filmmaking, and we’re probably not doing it with fewer people. In my experience it’s almost always more.
I work on a show with virtual production, and there’s more people than ever working on it. The level of specificity on it is higher than ever before. If the argument is that you can do anything in Unreal, that means you’re making more choices. We’ve got the Unreal team, we’ve got the live action camera team, we’ve got virtual sets and live action sets blended together. Then you have the visual effects, the editing and the whole post production process. You need a lot of people to complete the task of forming the finished project.
What is different on virtual productions is that you have a lot more creative prep upfront. We spend three or four months generating the virtual worlds before we start shooting them. Sometimes the script only provides rough details on the place and the context. I could see a future AI-generative production being entirely prep, possibly doing motion capture along the way, or mixing it with the live action component. There’s a section of the audience that might be turned off by knowing that something’s generative, even if it looks indistinguishable from live action. Then there’s the reality that live action capture and displays will also improve in resolution and fidelity, so I could see that for a time generative filmmaking may be only one tool or a blend of tools.
I could see everyone working together, building visual AI models and things that get specific to a movie, specific to a scene, specific to a shot. Once you have that model, it becomes easy to create any shot – but it could take you six months to get there. That’s how I could see it all blending together. The most exciting part is that I believe AI will make us more creative.
Kirill: Getting to “Star Trek: Strange New Worlds”, how was the process of taking something that has such a rich history of almost sixty years and bringing this fresh approach to it?
Benji: In terms of the contemporary nature of the canon of “Star Trek” that’s been established, almost all that credit goes to the writers. They’ve been doing so much research and cross referencing, and a lot of them have been on previous “Star Trek” shows. Glen Keenan, CSC came from Star Trek: Discovery and shot the pilot for Strange New Worlds and alot of Season 1, which set a great tone and inspired me on Season 2. That said, every episode of Strange New Worlds are different genres, and they’re made from scratch.
Having said that, probably 30% of any discussion we have about what we’re going to make in any given episode, any meeting, any conversation comes back to – how does that work within the Star Trek canon? How’s it been established? What are the rules of the transponder? What planets have they been to, and how does the ship work? Can we do this if they need a warm-up time before they go into warp? What color is the warp? So a good third of any idea that we talk about revolves around how it works with everything that’s been established.
Kirill: What is a virtual art department? How would you describe it in terms of what was before and how it is now, or how it was on this production?
Benji: The main difference, which may sound obvious, is that in the traditional art department discussion and collaborations had to do with a physical build, and this has to do with a non-physical build. When there’s really no limit, what is the level of creativity and specificity you can have?
The virtual art department (VAD) involves, among others, the production designer Jonathan Lee and his multiple art directors, the visual effects department, show runners Akiva Goldsman and Henry Alonso Meyers, and producing director Chris Fisher, our lighting designer Alida Keenleyside, creative heads, and the cinematographers. The interesting part of it is that the visual effects which historically came in post-production is now heavily involved in prep. We are getting to see finished worlds before our eyes and pointing a camera at it during production. And that is a great thing. In the virtual art department, we’re getting results from the visual effects department in conjunction with the art department during prep – months before we shoot.
We’re talking about building the entire planet. What’s the history of that? What about the people? Has this been established in “Star Trek” canon or is it new? No matter if it’s new or canon, we’re taking it to another level of specificity. Do the people use a bioluminescence to light their world, but they still haven’t figured out electricity? There’s all kinds of different divergent paths their evolution could have taken. It’s an interesting commentary on ourselves.
That’s an exciting thing, because everyone gets to be heard and have their instincts involved in the creating of these spaces.
Kirill: Do you find it useful to continue talking in terms of prep, production and post-production, or is it a continuous journey, if you will?
Benji: Every project is different. On a “regular” project, once we’ve shot it, the post-production takes over. On Strange New Worlds, I’m involved in color and a few other things, but because so much preparation has been made, there’s no feeling of wondering which direction it’s going to go. Sometimes there’s things that they decide they want to fix or change a little bit, but it’s almost always for the better. Almost everything is as expected, as the visual effects department does a great job. The journey is heavily prepped into production and for me, completes there except for some final touches. Not every project is like that. Sometimes it’s more heavily involved in post or there may be many more unknowns.
Kirill: There was, or maybe still is, an almost generational divide of shooting film vs digital. Do you see it becoming another “divide” of shooting practical vs virtual?
Benji: I want to clarify that we still build practical sets on every virtual stage. If you look at the measure of success, the more you’re relying on the practical set, with the background being a 3D extension, the more successful and realistic it feels. We’re still honoring and using the practical process. When you go virtual, it becomes almost an enhanced practical process.
Keep in mind that with the virtual capabilities, cameras keep getting higher resolution and higher fidelity. That is something that the virtual world has to chase. Live action is still the winner when it comes to the ability to move an audience. We talked about generative filmmaking earlier. Can it move an audience? I’m sure it will get to that point. But will live action be still more successful at it, with all the capabilities that it will be growing with?
Kirill: In this context, how do you define a successful production? People certainly see the seam between live action and generative AI output today, but let’s say that a few years down the line, that seam is gone and it’s indistinguishable. Do you look at the audience’s reaction, at the financial results, at how it fulfilled your artistic vision?
Benji: I’ve always used my own reaction to what I’m seeing as the gauge. I’m not in the business of doing the marketing and making sure it’s seen. I’m in the business of interpreting the story and giving it the proper visual representation.
I don’t know what other people think, and I can’t say an audience would or should feel a certain way. I’m an audience, because I love watching movies. That’s what inspired me to be in the business. So when I look at the screen of what we’re creating, I want to know if I feel something. There’s times I’m moved to tears on set watching a certain take. And even amidst all the distraction around the screen, outside the frame that I know is essentially manufactured, I can still be moved.
I think that’s the success, and that’s what an audience I would hope is looking for. When I’m moved emotionally, I don’t care how it’s made. I think that production will naturally go that direction where it’s not really important what tools you use. On our show, we’re using these currently expensive and, in a way, slow and sometimes difficult methods, because the result is more impactful to the audience. Some recent reviews even called my first episode (Season 2, Episode 2) one of the best episodes of Star Trek ever.
Kirill: Outside of the scope of interviews such as this, do you want the audience to think about how it was done, or do you want them to watch the story?
Benji: Definitely just the story. In the context of a “Star Trek” audience, there’s a lot of enthusiasm around how things are done, and the lore of this 60-year continuing story.
What I like about “Star Trek” is that it’s a commentary on where we humanity could be. The level of detail of the politics, the interactivity and the cohesive, albeit manufactured, science – is painting a picture of where we might go. That inspires an audience to want to know everything about it. In this history, we are picking up the story in a place where the challenges that humanity currently faces are more or less a thing of the past, and there are new challenges for us. It is also nice to work on a project that is not dystopian [laughs].
Kirill: If you look at the tools that were available to you on this season, are you excited about how they might evolve in the next few years? Without necessarily predicting how exactly, do you see them becoming a part of your arsenal?
Benji: Absolutely. Even between seasons of “Star Trek”, we’ve seen an improvement in capabilities in the Unreal 3D space, and in our own lighting tools.
If we’re talking about virtual production, something new we implemented on the second season was the ability to give a DMX address to a virtual light in the 3D world through our practical lighting board. That way our lighting board could affect lights that we physically set up on the stage to light our actors and the sets – and also virtual lights within the 3D world. For example, if we have a long hallway of lights that flickers on in a cascading sequence, we are able to time seamlessly between the virtual world into the physical world. That not only helps sell the effect that there is no blend, but it becomes a great way to open up the creativity and think of the virtual world as part of your set.
We also use a lot of big lighting “shapes” on the volume. Each one of those LEDs has its own individual address and output. We can crank them up to 100%, or any value, and create lighting from a section of the wall itself. This way we could make enormous soft lighting if we wanted to, or select tiny little areas on the ceiling. So we’re using the actual space as the lighting instrument a lot of the time. Into this next season, which is currently on hold, we were seeing more improvements in the 3D space of ways that we could enhance what we do.
To go back to something I said earlier, we’re almost never sacrificing time in order to do something the same as that we were before but quicker. Instead, we’re finding new capabilities, and spending the same time with the same number of people – or more people and possibly more resources – doing better, newer things. That’s how I see the trend going. You’ve got an enormous hunger for TV shows than ever [laughs], and the bar is raised from a visual quality standpoint that the audience expects. It keeps having to increase.
Kirill: This Star Trek universe started about 60 years ago. Let’s say it has another 60 years in it, and a whole new generation of viewers. Do you want to worry about how what you do today will be seen in 2083?
Benji: I think “Star Trek” as a whole is one of the only examples of this longevity, and it’s because of the story. They were using what we would call completely antiquated techniques to shoot the original series. So if you look at it from a visual quality standpoint, does it hold up? I would say inevitably no, it won’t.
So I’m not concerned with how long that will last, as much as how well the story was portrayed, and what are the instincts of the filmmakers. Those stand the test of time. Some of the earliest films ever made are still considered some of the best, alongside more contemporary films. It’s the depth of insight and the ability to give the audience a feeling that they’ve lived longer than two hours in that time span. If you’re successful, the audience will want to have that again.
Kirill: Would you say that the need to tell and listen to stories is biologically programmed in us?
Benji: It’s interesting to use the word “biological”, because this immediately brings up the context of AI which is the non-human counterpart.
I would say, yes, this is something that cannot be removed from humans. When we say “story” or “storytelling”, it can trigger certain thoughts about what that is. For a long time, my reaction to the word “story” was that it was a human created phenomena, something that is not inherent. I thought that it’s something that we’ve decided that we should do.
But I’ve come full circle on that. It is so much ingrained in a part of our communication, that we’re mimicking the life we feel and re-presenting it in a more concise way. That’s what I mean when I say that we are able to give the audience more than two hours worth of life they’ve lived in less time. In a way, storytelling is economizing the feeling of living longer into a more succinct time.
Storytelling to me has become the skill set of talking about what it’s like to live life. So, I do feel that it is biologically programmed or inherent to our existence, that you could call each moment in our lives its own story. If I used a different word for storytelling, I would call it “life re-presentation”, and you could maybe see it that way.
Kirill: Now that you are maybe a little bit older and wiser compared to when you started, if you had a time machine and you could jump back to when you were just starting out and you could give yourself one advice, what would you want to tell your younger self?
Benji: There’s so many things, and it’s hard to pick one. Probably to enjoy the process of “climbing the ladder”. It’s such a competitive industry that at some points, when I felt like I had so far to go, I was worried that I wouldn’t get there. Everyone is on a journey in their creative career, and you have to think what is it that you’re truly after? I was never trying to achieve a certain scope or budget. I was interested in projects that I would get continually more excited about. In a way, it was almost like I was chasing the increasing quality of scripts.
I don’t think I did anything differently because of that yearning to climb and to progress. But I definitely got stressed out along the way, because I felt the pressure that was self-imposed. So in a way, it would be to take my own advice and be relentlessly positive, but embody it even more. If you’re telling stories in any way, that’s really exciting.
And here I’d like to thank Benji Bakshi for taking the time to talk with me about the art and craft of cinematography, and Ashley Abshire for making this interview happen. “Star Trek: Strange New Worlds” is streaming now on Paramount+. Finally, if you want to know more about how films and TV shows are made, click here for additional in-depth interviews in this series.