Jeff Kaplan: World of Warcraft, Overwatch, Blizzard, and Future of Gaming | Lex Fridman Podcast #493
In a Nutshell
Jeff Kaplan recounts his journey from EverQuest addict and failed writer to Blizzard legend, leading World of Warcraft's quest revolution and Overwatch's hero-shooter success, while candidly detailing failures like Titan's cancellation due to hubris, scope creep, and corporate overreach. He shares leadership lessons—prioritize player fun over designer ego, small focused teams, hotfixable engineering, and saying no—plus the pain of leaving Blizzard amid financial pressures. Now at Kintsugiyama, he's building The Legend of California, an open-world multiplayer survival game set in 1800s Gold Rush California, emphasizing craft ownership, imperfection's beauty (kintsugi), and small studios' future over AI hype or big corp interference.
These notes were generated by AI and may contain inaccuracies.
There's three types of fun: fun for the player, fun for the designer, and fun for the computer. "Is it PvP?" "It's all PvP. In fact, Rust is the most PvP thing in all of PvP." "Rust players know what that means."
My whole career and my family are thanks to EverQuest, so I think I won the game. We're reading the forums, and the forums are just flaming us all the time. "There's lag on this server," and "Can't log into that ser-". That was our perspective of what was happening. When I showed up at that show, one of the most emotional things in my life. It was nothing but an outpouring of love. I had believed I would never work any place but Blizzard. I loved it. It was a part of who I was, and I felt I was a part of it, and I literally thought I would retire from the place. I never thought the day would come, and that was it. "How painful was it to say goodbye?" It broke me.
Meanwhile, as far as the outside world is concerned, you've disappeared off the face of the earth, but you were actually working on a game. Legendary game designer of World of Warcraft and Overwatch, two of the biggest, most influential games ever made. Genuinely one of the most amazing human beings I've ever met. In the many conversations, including while playing video games, always kind, thoughtful, hilarious, and still and forever a legit gamer. Quick to celebrate the incredible teams of creative minds he has worked with over the years. Blizzard has created some of the greatest games ever made: Warcraft, StarCraft, Diablo, WoW, Overwatch and more. Big thank you to Jeff, the entire Blizzard team, and every creative mind in the video game industry.
Super fun, inspiring, whirlwind conversation, full of memes, lulz, wisdom, emotional rollercoaster moments, and Blizzard video game lore. Jeff left Blizzard in 2021, and has been secretly working on a new video game called The Legend of California. Got a chance to play it with Jeff. Incredibly beautiful. Set in the 1800s Gold Rush era of California, it's an open world online multiplayer game, part adventure and action, part survival. Sometimes creating a feeling of loneliness and desperation, and sometimes just awe watching the sun rise over a beautiful landscape. Unlike any game that Jeff has ever worked on. Can't wait to play it. Wishlist it on Steam. Join the alpha later in March, and early access is on the way.
First a legendary video game player, in particular in EverQuest, before becoming a legendary video game designer on World of Warcraft and Overwatch. Wild journey from gamer to designer. When did you first fall in love with video games? Born in that golden era of coin-op. Remember the first time seeing Pac-Man with Uncle Ronnie, who kept feeding quarters. Too scared to play himself. At brother's graduation in Philadelphia, Asteroids machine in the lobby, one of the first coin-op machines played. Brother and I tried to get the high score, but had to go to bed early as little kids, and someone beat it by morning. Grew up in Southern California in the '80s, born in '72. Kid with skateboard BMX culture, riding two towns over to pizza parlors, liquor stores, and arcades, living in that coin-op phase. That was where the love started. Then Pong at friend's house, mind blowing playing on TV. Atari was big. Big one was Intellivision, because dad was executive recruiter, client Mattel, got discounts or free games. Brothers and I played endlessly. Comparison was always "Is this game close to what's in the arcades?"
Golden era. Big moment when NES came out with Super Mario, gaming at the next level. Warm, fuzzy memories. Played Super Mario for weeks with brothers, then friend showed secret stuff I didn't know existed. Suddenly, the world opened up more and games could be more. Big PC gaming push. Parents ran their own business, dad executive recruiter, bought an IBM with DOS before MS-DOS. Disappointed because other kids had Amiga or Commodore, better for gaming. Mom encouraged, bought Zork, Infocom word games where imagination takes you. Zork holds a place in heart few games will touch. Text-based game: type "Go west. Open mailbox." Power of imagination, why the book is always better than the movie. Starting to see creations of worlds you can navigate, step into, lose yourself, transported, living there.
Zork insanely popular, then Zork II and Zork III, a trilogy. In the '90s, CD-ROM games era, return to Zork with graphics shattered everything because the Zork in your head didn't exist anymore. Zork fantastic, might be open source now. Highly recommend Zork. On PC that worked on IBM, Ultima series by Richard Garriott, Lord British. Put himself in the game. World-building: Yew Forest, characters. First Ultima played was Ultima II, Ultima I before time. RPG group based PC game, worlds rich. Get on rocket ship in fantasy world fighting demons. Crazy stuff: bouncers in towns, merchants, could rob or kill Lord British, super hard. As jackass kid, spend time endlessly trying.
Ultima profound experience. Led to Ultima Online, legendary game, connected to EverQuest, starting to build massively multiplayer online video games. Big one was EverQuest, but Ultima Online predecessor, came before. Unfortunate time in grad school, missed Ultima Online. Stories craziest, funniest. Know somebody who learned to poison, poisoned apples, left on ground, someone feeds to horse, kills horse, steal stuff. Ultima Online earliest grief-based experiment, treating humans like ants in ant farm.
First online gaming defined by Quake, Doom, Duke Nukem. Started with Doom, LAN or modem connect. Mind-blowing seeing another entity that's a person on the other side, could be in another room or across town. Quake took to next level, systems refined. Quake community with websites, mods. Two castes: low ping bastards (LPBs) and the rest. Rolled into Quake matches on dial-up modem with 300 ping, greatest thing ever. Connecting with people, websites. Only gaming website read is Blue's News by Stephen Heaslip. Linked a story from there, teased for time machine. Through that site learned about EverQuest via programmer .plan updates. Guys like Carmack posted code, optimizations, personal life, Ferrari talk. id programmer Brian Hook leaving id to work at Verant (became Sony Online) on EverQuest. How does anybody leave id, greatest institution, for another game? Had to check out EverQuest.
Carmack, Quake: Early geniuses at id. Wouldn't be here without their breakthroughs. Gaming engines evolving, Wolf 3D first. Played Wolfenstein as 2D game, run around, dress as German, throw grenade. In 3D, transported, intimacy of first person, hands holding gun, to Nazi Germany as hero. Evolution to Doom, like Army of Darkness video game, "Give me the boom stick." Graphical advances, but how it played: smoothness, responsiveness, sharpness. Credit id, Carmack, Romero. Worked on FPS, wouldn't exist without them.
Gamer range incredible: legit first-person shooter gamer, loves MMO world, rich, exploratory. Technology stack for Quake or Wolfenstein 3D, threshold of realism to immerse. Same experience with Wolfenstein 2D to 3D, tears in eyes, ultra realistic, feeling like you were there. Storytelling via technology bringing 3D world to life. Master's degree in creative writing from NYU, wanted to be writer. Main influences: Kerouac, Hemingway, Salinger, Bukowski, Orwell. Drawn to storytelling in writing, aspect of human experience to put on paper.
Started as fan, inspired by reading, transported to different world or person, stories touch emotions, trigger feelings you didn't know you had. Big challenge: putting yourself out there. Ego in saying after reading 1984 or Green Hills of Stranglethorn, now try to write something people read. Giant leap of faith, part ego, part masochistic. People who create can't help it, wired that way. Admire Dickinson writing poems left in drawer.
Franz Kafka stories never published, asked to be destroyed, friend ignored. James Joyce at 18-19 declared greatest writer of 20th century, turned out to be. Ego necessary. Threw away everything in dumpster: 20 volumes handwritten notes, journals, manuscripts. Necessary, broken and defeated by failure, throwing in towel like boxer. Received over 170 rejection letters in one year submitting stories. Hard. Moved from New York, left great writer friends group. Writing lonely, but writers support, give honest feedback. Moved back to California for a girl, broken up in two months. Knew it was epic mistake closing studio apartment door in New York. Alone in California, structured life to write eight hours a day, job not hobby.
Fortunate, dad hired as research associate, calling generating name lists, took East Coast assignments to start at 5:00 AM, created space to write. Dog named Jack, Jack Russell Terrier, named after Jack Kerouac. Mid to late '90s, sent printed manuscripts to literary magazines with return self-address stamped envelope, expensive. Paid for rejection letters. Hoped for editor note, like "Glimmer Train" saying showing promise, cling to it. Soul crushing. Stuck, became isolated, left writing friends, introversion, breakup, all writing. Deep heavy depression, drank too much, alcohol problem. No magic rejection, moment said this will destroy me. Universe of editors telling time to give up. Stopped. Closing a door required for another to open. Hardest to walk away, invested self, dreams, conception of self, becoming isolated, suffering. Way out is dumpster, cut it off.
Hindsight, ended up okay. Young in late teens early 20s, pressure, adults don't help. Every time run into younger nephew
Society pressures young people with questions like "What's your major? What are you gonna do? What's your career?" This is bullshit when you're lost, especially in your 20s. The point is to move through the world to find what sparks your heart, your passion, what you're meant to be on this earth for. That's a real hero's journey. Adults often stop searching, settling comfortably, forgetting the suffering and uncertainty young people face. There's confusion and insane pressure to have it figured out.
Focus on what you want to do, not what you want to be. Society asks "Do you want to be an astronaut? Firefighter? Writer? Game maker?" People get lost performing as a fake actor in that role. Instead, when off the clock, away from pressure like at Thanksgiving, ask: How do you spend your time? What makes you happy? What brings fulfillment? Through those paths, you'll find what you become. Do the thing that brings joy moment by moment.
After giving up writing, structured eight-hour writing chunks ended. With free time alone with dog Jack, poured it into EverQuest (released 1999). Friend Victor, one of few who played computer games despite stigma, recommended it. Brian Hook worked on it. Logging in transported to world of Norrath as halfling rogue. Gorgeous world, mechanics of adventure, killing for experience, better loot for progression were amazing. Lived for next login, much escapism, not all healthy. Quit after three years: /played command showed 272 days (~6,000 hours).
EverQuest: Long-running 3D fantasy MMORPG set in Norrath, first released March 1999. Thousands create characters, group, explore persistent shared world. Foundational MMORPG defining raid content, guild systems, 3D online worlds. Raiding: 30+ people conquer what one couldn't. Needed Uber Guild. Played unguilded or rogues-only role-playing guild mostly. At Level 50 (top level), joined Legacy of Steel, top guild on The Nameless Server. Thrill of 30 people coordinating to beat Nagafen (fire dragon) or Vox (frost dragon). Screamed alone in room, cheered wins, depressed on losses. High highs, low lows, did everything right.
Big leap from proud lone warrior to Uber Guild member, then leader of Legacy of Steel. Organizing in EverQuest like herding cats: loot-motivated, guild-success driven, lonely seekers. Much depression in community; games provide escapism, healthy in moderation but abusive when avoiding real life. Cycle: camaraderie relieves, but withdrawal worsens need for game fix. Psychologically complex. Coordinated diverse people (all talking shit in late '90s gaming culture) from all walks: repeat instructions 200 times, e.g., "port from North Ro" for hours to get 20% turnout.
Joined as bright-eyed rogue, few in guild, wanted to help, admired leaders. Guild leader disappeared suddenly (common in EverQuest, no notice). Made assistant guild leader, raid leader, officer by Ariel (blonde wood elf warrior refusing helmet). Running raids natural: first leadership experience learning motivation, discipline, inspiration, when to force vs. encourage. Safety net of real leader. Ariel too busy, handed over guild fully.
Most anonymous via usernames, no real-life info due to stigma; assumed male, surprise if female. Met wife in EverQuest. Critics say threw years away, can't win such games; response: entire career and family thanks to it, so won the game.
Hundreds of hours, some happiest of life despite feeling low during. All-nighters immersed, sun coming up, many Blizzard games. Video games like books take you on journeys but had stigma; felt like secret dark thing, heroin, usually in dark, secretive. Not mainstream. /played on TV watching would exceed 6,000 hours (20 years ago). Hard to say goodbye; tried quitting breaks, thinking done for good, but returned. Final quit easy, psychologically done after Shadows of Luclin expansion (didn't resonate like Scars of Velious). Got Blizzard job; obsessive, shifted all energy there.
Tigole (username): Legendary gamer/troll giving edgy feedback to devs, e.g., "pull heads out of asses," but specific improvements. Loved game, cared. Loved/respected for that.
First guild leader Dread (founder) left, guild listless. Ariel stepped up, made assistant. Around Half-Life 1/Duke Nukem, shipped editors on CD; made levels, published in UK magazines. Ariel requested Half-Life levels, revealed identity: Rob Pardo, Blizzard designer, Warcraft III lead. Lived in Pasadena (mentioned Rose Parade on site), Irvine hour away. Guildmates also Blizzard: Scott Mercer (Dalomin, enchanter), Roman Kenny (psychotic wizard). Mailed levels on CD (internet upload took 12 hours). Never played StarCraft, Diablo, Warcraft; selective due to cost. Blizzard building legendary status, seemingly doesn't miss.
Drove wary (internet stranger danger). Lunch with Rob Pardo (Ariel), Scott Mercer, Roman Kenny. Coming out moment: talked games openly, no stigma/embarrassment. Felt like himself first time, true happiness surrounded by game lovers. More lunches. Met Barfa (troll warrior): saved in The Hole dungeon with expensive teleport potion (rogue could stealth out). Barfa = Allen Adham, Blizzard founder, head with Mike Morhaime, Frank Pearce. Lunches were subtle six-month interview.
Rob logged in: check Blizzard job site. Posted World of Warcraft associate quest designer job, wanted creative writing degree (his background). Applied wholeheartedly. Hardcore six-month process; Blizzard rarely hired external designers (promoted from QA, tech support, internal). Met Chris Metzen (inspirational, creative), paired with quiet Kevin Jordan (early WoW class/PvP designer). Lunch at Italian place; Chris bought cigarettes, cursed freely. Felt around "his people," unlike corporate culture.
Jeff Kaplan's last interview at Blizzard was with Alan, Rob, and programmer Bob Fitch, one of the first five developers at Blizzard. They took him to an ARCO station with a Jack in the Box combo. He thought, "These guys just brought me to a Jack in the Box that's in an ARCO station. I need to work here. These are my people. This is where I belong."
Through his journey, Kaplan faced low points including depression. He started drinking heavily, wrestling with alcohol until his early 30s. He's proud of his sobriety. He would buy a bottle of Old Grand-Dad, drink it alone, and watch the Oscars, feeling like a failure compared to the polished celebrities. He struggled with trying not to drink but drinking to feel better. His parents were supportive. Therapy was extremely helpful; find the right therapist who helps you get out of your rut healthily. He tried antidepressants but hated taking pills and feeling altered. The hardest thing was electroconvulsive therapy (ECT, shock therapy), which broke him out at a desperate low point when people worried about his wellbeing. It was an extreme pull-the-rip-cord moment. Starting at Blizzard also helped.
There was deep loneliness before meeting the Blizzard team, unlike in New York with his writer's group. Seeing the guild-like team was everything he needed. As an introvert, he believes introverts need people more but struggle to engage healthily and connect.
Creative writing attracted him due to its solitude and lack of collaboration. Film industry internships at Universal Pictures showed unhealthy collaboration and a caste system where low-caste people were treated horribly. Games were different: flat structure. Mike and Allen treated an associate game designer like an equal. Comradery and strong work ethic on meaningful work pulled him out.
Lex reads Kaplan's post titled "If you want something done right." Kaplan accepted position as associate game designer at Blizzard, designing quests for World of Warcraft (MMORPG based on Warcraft series) and contributing to endgame content. He shares to assure fans one of them is inside looking out for players. He criticizes tedious key camps, obvious time sinks without story or narrative, quests rewarding luck over skill, and out-of-sync rewards. Hopes his involvement comforts MMORPG fans. Final paragraph: excuse lapses in site updates; NPCs need to learn words like "kaksagur" and "mo'fucker." Already in trouble with the boss.
Boss Allen was understanding; they knew what they were hiring. Post embarrasses him now due to ego and not knowing what he didn't know. Passion from grinding EverQuest gave insight into compelling experiences, but he didn't grasp the hard work, uncertainty, trade-offs, and how designs change in practice with millions of players. Ego is needed initially for guts to try.
Blizzard office on UC Irvine campus research park. Three-fourths of building was Blizzard (<200 people total in May 2002, including Blizzard North in San Mateo). Blizzard South (Irvine): Team 1 (RTS: StarCraft, Warcraft II, working on Warcraft III) revered; Team 2 (red-headed stepchild) making World of Warcraft after scrapping Nomad. Key designers: Eric Dodds (later Hearthstone director), Ben Brode, Eric, Kevin Jordan. Tech lead John Cash (from id Software, Quake III skin). Mutual fanboy moment as EverQuest players. Few producers.
- Engineers/Programmers: write code.
- Art team: 3D modeling, characters, environments, animation, tech art.
- Game design: create experience, systems, content (quests, story). Spectrum: systems designers (loot drops, percentages), content designers (quests, storylines), narrative designers, jack-of-all-trades.
- Production: project management, varies by company (boss, designer, or manager).
- Sound/Audio: sound designers, composers. Audio and lighting are unnoticed until missing; elevate experience.
Lighting under art, partners with graphics programmers. Engineering spectrum: architectural (server-client, networking), gameplay-focused, audio programmers, graphics programmers (shaders for art vision). Creative tension: artists (emotional, creative) vs. programmers (logical, organized). Middle: sound designers, game designers, producers. Brilliance in uniting vastly different talents toward shared game vision.
Small teams avoid compartmentalization; everyone has a loud voice. In 10-person incubation, all in every decision room (networking, art style). Bigger teams alienate disciplines, leading to vilification (e.g., "artists just don't get it"). On small teams, know everyone, no stereotyping. Harmful to demean others' life's work.
Assume top talent: treat prop artist as best in industry, listen deeply. Instead of defending your idea, consider why the expert suggests alternative.
Insecurity manifests differently; use it positively. As young lead on WoW, felt need to be right. Pardo advised: always listen and try to make others' ideas work. Shifted to core principle: develop ideas ("how could we make it work?"); give credit when succeeds, elevating team members.
Kaplan always humble, gives credit to team for everything.
Jeff Kaplan is known for introducing himself as "Hi, I'm Jeff from the Overwatch team," always crediting the team.
World of Warcraft (WoW) is a massively multiplayer online RPG where you create a character, level it up doing quests and dungeons, progress gear and power in an open fantasy world called Azeroth. Basic gameplay: move, use abilities from action bar, follow quests, learn combat rotation fitting your class. Choose race (appearance, starting zone, small racial bonuses), class (how you fight, role in groups).
"World of Warcraft, first of all, more than anything, is a world. Like, it's a world that you can live in with real other people, and everybody's kinda living out their fantasy."
Chris Metzen, creative director on WoW, called "the heart and soul of Blizzard" by Allen Adham (Blizzard founder). WoW embodies Chris Metzen's imagination.
"The lead character of World of Warcraft is the world."
World is exciting, dangerous, comfortable, uncomfortable, gorgeous, massive (half hour to cross end to end). Divided into two warring factions: Horde and Alliance—decision championed by Allen Adham.
Early days had strong division: pick a side, only group with your kind. Players get Horde/Alliance tattoos; deep identity (e.g., rejecting friends on opposing faction server like Tichondrius).
Horde-Alliance split controversial vs. EverQuest (mixed race grouping). Pardo and Jeff Kaplan (from EverQuest) argued against it. Daily lunches with Allen Adham, Rob (finishing Warcraft III), Bob, Fitch debating core design.
Allen Adham inspired by Dark Age of Camelot (three factions, instant teams, no loners). Rob and Jeff argued, but pre-beta, Allen retired (9-12 months before WoW shipped) to run hedge fund, poker, finance. Rob took over as lead designer, honored Allen's vision, implemented Horde-Alliance. Team went all in.
Early WoW team: Mark Kern (running team), Shane Dabiri (legendary developer), Bill Petris (art director), Metzen (cool big brother figure). Newbies + pulls from other teams (e.g., Quake community for dungeons, built as Quake levels in QERadiant, later proprietary engine). Hodgepodge like Bad News Bears.
Jeff Kaplan: EverQuest player hired to design quests. Success partly because team didn't know what they were doing—unlike Titan (post-WoW MMO attempt, best talent, failed due to over-expertise).
Blizzard (successful with StarCraft, Diablo, Warcraft) + yayhoos challenging Sony Online's EverQuest II (announced, highly anticipated) and Star Wars Galaxies (Raph Koster from Ultima Online, brilliant designer lecturing at GDC). Felt doomed to compete.
Small hodgepodge team worked crazy hours voluntarily, heart in it. Games industry crunch reputation: now banned/frowned upon (articles if overtime). Back then, insane hours (Jeff's longest: 30 hours straight gold mastering Warcraft III July 3, 2002—late June, additional testing cinematic crash reproduction).
Post-May 2002 E3 WoW prep, all hands on Warcraft III. Watched same cinematic 30 hours; surreal seeing team trickle out (families early), return refreshed while he stayed.
WoW crunched too; slipped despite Mark Kern's "crunch early to avoid late crunch" (believed genuine, but slips inevitable). Death march: Newport Rib, Panda Express catering stained carpets, falling asleep on couches, unhealthy diet.
Jeff never blamed execs—self-driven (Memorial Day with beach sand, Christmas work). Complicated: fulfilling growth for many (esp. 20s), incredible in retrospect despite pain, even unshipped projects like Titan.
Reflections: problematic when forced/unnecessary (e.g., "if anybody stays, we all stay," execs on 40h pushing others wrong/immoral). Personal: 10h minimum day, weekend work for passion/fulfillment. Fondest memories from WoW crunches; doesn't impose.
Pre-WoW MMOs like EverQuest: stand in one spot killing monsters hours. WoW: revolutionary quest-driven leveling—story-driven guide through world, levels character immersively.
Allen Adham (lead designer) wanted quests central (EverQuest quests obscure, website-hidden). Hired Jeff Kaplan and Pat Nagle (ex-HR/Facilities, handled apps/toilets) as quest designers, shared office.
Eric Dodds helped interface/NPC interaction. Whiteboard: match EverQuest's quests (3-4 expansions). Split zones: Pat Elwynn Forest (human start), Jeff Westfall (next). Met Chris Metzen for zone stories aligning gameplay.
Team playtest (Elwynn/Westfall, 12-20 quests each): non-MMO players (shooters, StarCraft) ran out fast, expected quests throughout. Oh shit moment: underestimated quests needed.
Philosophy: design along path of least resistance. EverQuest: grind easiest mobs in place. WoW: overload XP into quests to move through world, see everything, tell stories—fundamental genre change.
WoW success (tens of millions vs. EverQuest hundreds of thousands): play as single-player (MMO massive via others, but no forced interaction off-putting). Directed gameplay optional-feeling but core (vs. tight like Last of Us log-duck, open like Minecraft).
Revolutionary: enabled hundreds/thousands hours. Design loop (accept quest → kill 10 rats → turn in → level): fine-tune for max fun/max people extremely difficult.
Lack self-awareness of gamer type (wish vs. actual). Exceptional design needs playing tons (great/shitty games teach; analyze fun/unfun, systems, tuning, interface, polish). Watch others (wife, kids): differences, frustrations, misses. Raw honesty essential.
Jeff analyzes his gamer type: enjoys cow level and loot. Questions why loot is fun, like opening a chest. Core enjoyments: loot and walking beautiful worlds with nice music.
As game designer, a quack psychologist. Motivators: extrinsic (e.g., loot) and intrinsic. Players shift between them. Loot boxes in Overwatch masterfully designed by game designer: visceral opening, sound, graphics, animation as satisfying. Taps lizard brain (feels good) and spreadsheety part (is it upgrade?). Great designers tap both.
Studying writing: elements of fiction (plot, character development, setting, theme). No textbook for game design. Considers elements of fun: differ per human.
- Progression: investment recognized (leveling, gold).
- Mastery: pure skill.
- Creativity and customization (aesthetic like black curly hair, earring; build like whirlwind barbarian).
Designers tap these. Players drift: Jeff content-motivated (world exploration), detours to loot-motivated (need armor/sword for dragon/demon), or creativity (counter boss abilities like area of effect curse). Cycles tap different fun elements to discover world.
Wears many gamer hats: loves RPG/MMORPG, big shooter fan including ultra-realistic like Call of Duty.
Fun in shooters: skill/mastery (clicking heads), intimacy of first-person camera (look at hands, visceral test: click fast enough). Prefers first-person purity over third-person trend (PUBG, Fortnite, ARC Raiders). PvP: opponent coming at you.
PvP: player vs player (combative, e.g., Lex vs Jeff). PvE: player vs environment (computer-controlled, e.g., dragons). Can be multiple players: 5v5/6v6 PvP, raids (large groups vs AI).
Single player: play alone, not networked (e.g., Story of Seasons on Switch: personal farm/town).
Single player difficult/expensive: handcrafted (e.g., Uncharted, Last of Us by Naughty Dog).
Co-op: cooperative, sharing exact experience with known people (e.g., Left 4 Dead: 4 players fight zombie hordes to safe room; Diablo IV).
Multiplayer: strangers in same world, opposed/co-op goals or pass by.
Massively multiplayer online (MMO): thousands of players, huge worlds.
Couch co-op: two people together (remote or local). Tricky to design, fulfilling (e.g., Diablo IV). Predates couches: arcade games (two joysticks), Gauntlet (four), NES/Sega Genesis NHL 93 (up to 4, loser loses controller), Mortal Kombat (house rules: cycle players).
Magic in multiplayer sharing. Jeff never made single player game; admires but loves multiplayer.
Game director on multiplayer: like movie director setting up actors/sets/scripts, then leaving room. Watch players create experiences.
Anti-shitty story guy (loves story). Magical stories: player stories (e.g., gave Barfa potion, met in real life; loot goblin detour with another player). Better than written stories.
Exceptional writing: Naughty Dog, Valve (Half-Life 2 by Marc Laidlaw; Portal/Portal 2 by Erik Wolpaw), Rockstar (Red Dead Redemption 2): mastery of design/narrative + player stories (e.g., confused controls, accidentally shotgun town guy saying "Howdy", becomes wanted, train takes posse).
Red Dead: touching Arthur Morgan story + goofy player stories. GTA: radio adds world (humor, color, depth, heartbreak).
Moment in Red Dead: Daniel Lanois "That's the Way It Is" while riding back, quotes from Dutch/Arthur, Woody Jackson score transitions to lyrical. Hits emotionally: top of pyramid is art.
Investment leads to tears.
Image: zone flow for original World of Warcraft. Started from Chris's inspiration/world. Whiteboards: Chris drew gorgeous maps, told stories (e.g., Eastern Kingdoms, Dwarven Lands, Wetlands, Khaz Modan, Elwynn Forest, Westfall, Defias Brotherhood, Deadmines).
Capture spirit in heads. E.g., Burning Steppes: scary (lava, dragons), end-game zone. Starting areas: free flight path Ironforge (dwarves) - Stormwind (humans). Newbies fly over Burning Steppes/Searing Gorge, see scary places, plants seed.
Green Hills of Stranglethorn: emotional value, unanimously hated as shittiest/annoying. Wrote short story homage to Hemingway (quest giver Hemet Nesingwary = rearranged letters; another Kerouac mix-up). Wrote over winter break alone in office.
Ant farm designer (bad: play God, players as ants). Junior designer hubris. Sid Meier quote: three fun types (player, designer, computer) - prioritize player fun.
Story divided into pages scattered in Stranglethorn Vale, looted from any creature (bridging quest, overlap hubs). Like McDonald's Monopoly: collect all. Interface issues: pages don't stack, no container, limited bag space - fills inventory with duplicates (e.g., three page 5s).
Puppeteering: Stranglethorn chat trades ("page six? page three?"), intended social but annoyed players. Many skipped, used Auction House.
Learned from mistake, admitted failure publicly: opened Blizzard to self-criticism, challenged better quests.
WoW: biggest game (popularity, players, revenue, acclaim). Jeff game director for Wrath of the Lich King (greatest expansion).
Blizzard polish: studio-wide culture, no satisfaction with bugs (all games have them; fix quickly/urgently). Developers play as much as players.
Tactical: best QA (quality assurance) in industry - passionate gamers (many aspire developers). Tight dev-QA relationship: QA sat with teams, direct contact (home numbers), shatter chain-of-command fears.
Systematic: test plans, regression testing, compatibility testing (every config: cards, machines, motherboards). Not just play - methodical.
Developer interaction with QA revealed certain
On Overwatch, QA specialists were esports-level shooter players who could snipe from 100 meters and detect a frame of input delay. Developers trusted QA by pairing them with engineers to verify issues like input lag.
On World of Warcraft, QA built a full raid team to test raids, identifying bugs like dragons sinking through the world and crashing the game. Valuing QA meant asking their opinions: "10 million people are gonna see this. Your opinion matters. Are you having fun?"
Blizzard engineering architected games to be hotfixable—server patches without client downtime. Patches require downloads, but hotfixes allow quick responses to crashes or exploits, like disabling a hero in Overwatch within half an hour to keep players engaged.
Players feel developer love through quick fixes for balance issues, like everyone playing barbarian whirlwind. This craftsmanship shows in bug fixes, balance, and delightful secrets like the cow level.
Jeff Kaplan was game director of WoW when it was one of the biggest games. Joined as entry-level at $35K/year. After launch, Allen Adham left before beta, Rob Pardo took over then left for StarCraft II, putting Jeff (Senior Game Designer) and Tom Chilton in charge in early 2005.
Tom Chilton came from Ultima Online, had more experience. Servers were overwhelmed, world on fire. Jeff thought the chaos was normal for game development—working insane hours, then playing with wife, creative meetings with Chris Metzen, debating button placement for hours.
Handled disasters like GMs calling home at 3 AM about exploits, e.g., faction tokens in Stranglethorn Vale. GM spawned Guardians of Blizzard (giant infernals that instant-kill) to stop exploits. Pre-cellphone landline calls in hushed tones while wife slept.
Running WoW Live with Tom and team taught how to operate the game: patching, content types, emergencies, customer support. Debated launcher—originally for anti-cheat, tried removing it, but Mike Morhaime insisted: "There's no better way for us to talk to our players." Launcher became Battle.net.
For Burning Crusade, Jeff and Tom were leads. Post-WoW ship, two exoduses: one group quit to form Carbine (made WildStar in 10 years), unhappy despite success—possibly over promised compensation. Games peak in imagination at pitch, then reality sacrifices vision over 5 years (Overwatch 2.5-3 years), leading to disenfranchisement.
Team demoralized after WoW shipped: servers overloaded, unexpected success, hiring database programmers. Multiple exoduses—Carbine took animators/programmers; Red 5 took team leader Mark Kern and art director Bill Petras.
Patches made with one animator who later left, art team gone, morale low on Team Two. Felt like failure.
BlizzCon 2005 inspired by EverQuest meetups. Rob Pardo pushed: "Blizzard's bigger than that. We're not just one game." Included StarCraft Ghost, Frozen Throne, StarCraft II motion, Blizzard North.
Initial announcement got crickets; forums flamed lag/server issues. Mike Morhaime put tickets on launcher—sold out instantly. At show, outpouring of love contrasted online hate. Passionate fans online express harshly for attention; in person, full of love and nerding out.
Online culture favors hyperbole, especially negative, for engagement. Positive gets mocked; safer to criticize. Jeff contributed to early edgy posts on forums, social media before named.
Younger people afraid to show awe—default to "everything is stupid" for safety. Act of courage to say "Holy shit, that's awesome."
Advice: Be vulnerable, say you love things, especially young people shaping internet's future. Support creators—they weather firehose of criticism. 100 compliments forgotten, one insult lingers years.
Worries social Darwinism shuts out brilliance, e.g., Jay Wilson (Diablo III director, great designer) retired from games due to heat, wrote novels, now returning. Millions loved Diablo III, yet community drove him out 10-15 years—criminal.
Support young creators, allow them to suck/improve, take risks, be cringe. Comedians bomb but persist.
After Wrath of the Lich King, stepped down as WoW director to develop Titan, epic MMO to succeed WoW.
WoW seen as 5-year success; studio needed next MMO. Around 2005-2006, talk picked up during Burning Crusade. Rob Pardo started Titan ~2007 with cross-company group. Ambitious project, first team members joined.
Building a new engine from scratch. First team member was John LaFleur, stellar game programmer. Engine, which failed for Titan, became the engine for Overwatch, a great success. Game set in future Earth; players as secret agents with day jobs by day and cool secret agent stuff by night. Secret agent stuff was first-person shooter with over-the-top abilities like in Overwatch. By day, run businesses, influenced by Animal Crossing, Harvest Moon, Sims. Brilliant game designer and director Matt Brown, creative director on Sims, joined team. Vision included daytime business, house building, living in neighborhoods. Technically, unlike World of Warcraft's sharded servers (5,000 people per server), Titan aimed for one server, one massive world on future Earth. Built Bay City (San Francisco), Hollywood, all of California between, Cairo, London. Included full GTA-style driving.
Gargantuan undertaking: brand new engine, team, IP, setting. Wrestled with world-building rules, e.g., aliens or not. Development started 2007 ideation, really going late 2007. Shipped Wrath of the Lich King on Election Day, Obama's election, Kaplan's birthday, then left WoW team to join Titan. Shut down in 2013. Painful process; by deep into 2009 knew game couldn't ship. By 2010, after trying to convince leadership, went to CEO Mike Morhaime: "You gotta shut us down. We're just gonna burn money."
Multifaceted failure due to leadership, including Kaplan. Failed on art, engineering, design. Some best art ever at Blizzard made for Titan, but no art cohesion—looked like 10 different games. Cost $83 million. Ideas easy (e.g., secret agent running flower shop by night); vision requires shepherding through inspiration, team belief, technological plan, design plan, art style, pragmatic production.
Blizzard hubris post-WoW: "We can do no wrong." Anticipatory hiring: hired 70 environmental artists in one year before defining story, art style, tech constraints. Top talent on visas, unaware of project (e.g., guy from Belgium shocked on first day). Right way: smallest group, prove idea with any tech/art, then expand. Idle hands problem: huge brilliant team without roadmap led to busy work (e.g., props for Chinatown or Hollywood). Engine broken; playtests required "stop checking in." Technical artist Dylan Jones logged 20 hours uptime in 40-hour week in Titan Edit (TED), now Overwatch tool. Burning cash, demoralized team like fighter pilots not flying.
Overwatch: masterful execution, short timescale, small team, clear vision. Titan said yes to everything; Overwatch said no—deep focus. Creative leadership: pushing (inspire when not thinking big) or pulling (focus to ship). "The best feature we can add for the player is shipping."
Titan canceled, team to disband. Kaplan fought to keep core together; Mike Morhaime gave six weeks for new pitch. Criteria: ship within two years (insane for Blizzard), potential for World of Warcraft-like revenue (only WoW had it). Ignored revenue as Monopoly money. Team shrunk: 140+ to small group (most to Heroes of the Storm, D3 expansion, WoW, Hearthstone); temp loans; engineers mothballed Titan, deconstructed engine. Three ideas, two weeks each, all-in: fully commit then discard.
StarCraft MMO, StarCraft Frontiers. Lone space prospector (old Chris Metzen drawing), not space marine armies. Ground-level: Indiana Jones in space, finding monsters in dungeons while prospecting minerals amid Zerg/Protoss/Terran battles. Concept art by Arnold Tsang and Peter Lee: prospector smoking cigar, foot on Hydralisk skull, Medivac background—later became McCree. Full pitch: world design, class design, progression, expansions/live content. Still 5 years with 150-200 people, not 40 in 2 years. Idea, not vision—lacked path to reality. Endless scope creep (e.g., space flight adds years).
Brainstorm: "What's fucking amazing?" Blizzard value: "What is the fantasy?" Directors (tech, art, game) scope to reality, not producers/execs. Tension: think big, then sculpt down over days (10 working days per pitch). Avoid prop artist spending weeks on one crate; need 200, do in hours.
Chris Metzen vision: planet on universe edge like Mos Eisley—seedy, criminal, freaks/aliens/traders/smugglers/diplomats, neutral meeting place. World IP-driven, single planet. Allows imagination for off-world origins/expansions (e.g., Green People planet). Design meetings: class vs. skill-based (trade-offs). Jeff Goodman (original WoW encounter designer, e.g., Onyxia, many raid bosses): "I wish instead of six classes, 50 classes, each with one or two interesting things." Stuck with Kaplan.
Browsed Titan art directory. Arnold Tsang characters unmatched. Pulled old Titan Jumper class, who could teleport forward and
The Jumper could teleport forward, rewind time, and come back. The Jumper used dual-wield pistols designed after dual G18s from Modern Warfare 2. Every version of the Titan Jumper was distilled into dual-wield pistols, blink, recall, and time bomb. Arnold Tsang's art was used, and Arnold suggested making her British named Tracer. This was the origin of Overwatch. Pragmatically, Geoff Goodman and Arnold Tsang were on the team, so the design played to their strengths for a realistic two-year development.
Titan classes were reviewed with The Gunjack, who became Reaper. The Ranger split into 76 and Bastion.
Overwatch features a large number of heroes with distinct looks, skills, and personalities, like Lena Oxton (Tracer), rather than few classes with many skills. Backstory seeps in indirectly throughout the game.
A seven-page deck called "Monetized Shooter" featured League of Legends plus Team Fortress 2 logos and six sloppily designed heroes. While working on CrossWorld, Ray Gresko (wrote Dark Forces engine, production director on Diablo III, helped kill Titan) saw it and told Jeff to show Chris Metzen. Metzen approved immediately. Arnold's art was used. Ray and producer Matt Hawley insisted on pitching it the next morning despite focus on CrossWorld. Matt Hawley refused to present a deck titled "Monetized Shooter," so it was renamed Overwatch on the spot. The name came from a Titan whiteboard vote for favorite names where Overwatch (a police group) got the most votes, despite rigged process—naming it was a high-five to the team and middle finger to fake democracy.
Jeff's deck was gray with black writing and default PowerPoint blue shapes, dragging in Arnold's art. Jeremy Craig (now game directing at Bonfire) refined it into a gorgeous pitch. It went through Blizzard production/game directors, executives, then Activision.
To address schedule, the deck used crawl, walk, run. Titan failed by trying to run (next World of Warcraft) without Warcraft I-III foundation. Crawl establishes a universe worth caring about. Two-year timeframe allowed no compelling PvE, so dates were aspirational to avoid cancellation. Also addressed mobile strategy with a Photoshop of Arnold's art on a tablet. Crawl: shooter with heroes. Walk: PvE co-op. Run: big MMO if world succeeds.
Overwatch world is set on Earth in a bright, hopeful near-term future—a "future worth fighting for." Avoided dark, gritty post-apocalyptic or ultra-realistic like Call of Duty. Locations chosen for fun: Santorini, aspirational Oasis (bright version of Iraq). Avoid cargo container mazes; create pleasing, non-oppressive environments as world tour of great places.
- Tracer: OG, cornerstone, on box art. Simple mechanics: dual pistols, blink, recall. First recall use is mind-blowing.
- McCree: Simplicity in design. Six-shooter feels phenomenal (Mike Heiberg designed High Noon ultimate). Tumbleweed rolls during ultimate.
- Reinhardt (Jeff's main): Geoff Goodman design. Shield (shoot-through, now copied everywhere) puts players in protector mode. Rocket Hammer, charge ability inspired by Left 4 Dead 2's Charger (a commit ability). Great Reins separate from bad by not charging to death.
PvP 6v6 with roles: tank, support, damage. Initially no enforcement, later 2 per role for fun. Heroes have lore, personality, community.
Huge influence from Team Fortress 2 (Quake mod, Team Fortress Classic with Half-Life 1, TF2 in 2007). Arcade/arena shooter style: fast movement, epic weapons, low TTK (survivable, take hits unlike CoD/Counter-Strike headshots).
Most complex design/engineering tasks, thankless. Players want fair matches but really want to feel slightly better and win close games—zero-sum, can't architect. Losing streaks highlighted on Reddit, but winners don't post (e.g., 6 losses after 8 wins). Regret: too team-focused (no scoreboard, medals weaponized on losing team). Advice for hero shooters: downplay team, emphasize individual contribution—humans are selfish by nature.
- Quake: GOAT.
- Rust: Most hours logged. Open-world procedural island PvP, resets monthly. Players kill, loot, build/raid bases (online/offline raiding). High highs require low lows (e.g., perfect base raided, back to nothing). Like WoW PvP but loot everything, reset to level 1.
- Call of Duty: CoD4/Modern Warfare 2 pinnacle, Black Ops third. Masterful gun feel (VFX, animation, modeling, sounds). Maps like Crash (CoD4), Terminal (MW2) studied for flow.
Overwatch 2 (PvE co-op shooter, walk phase) started in 2015, pitched pre-Overwatch 1 (2016) launch with Chris Metzen and Michael Chu for runway. Biggest career mistake as leader; two points of failure. First was having people on the game team
Team members who didn't like PvP or competitive shooters loved the Overwatch universe and wanted to play heroes in a PvE setting on their own terms. Despite Overwatch's success, they couldn't engage with it. Dev team was excited to ship PvP and immediately start PvE work.
Executives from Blizzard and Activision pressured for Overwatch 2, referencing slides with unrealistic dates like 2019. PowerPoint decks become etched in stone for executives. Dates were placeholders, similar to putting Tracer on a tablet for a mobile strategy.
In May 2016, during Rio Olympics, proposed Summer Games event (avoiding 'Olympics' to prevent lawsuits). Developers Mike Heiberg and Dave Adams created Lúcioball soccer mode with map and mechanics. Event was extremely successful, leading to Halloween event. Fans loved events.
Dev team subgroup wanted to focus on Overwatch 2 instead of live events, criticizing over-scoping patches. Advice: catch the wave and ride it; most games fall off quickly.
Coalition on team prioritized Overwatch 2 over live events. Executive pressure mounted. Correct path was more world events. Major derailment was Overwatch League.
Kaplan believed in Overwatch League: pitched regional teams, minimum player salaries, protections to make esports first-class. City-associated international teams for real competition.
Excitement led to over-marketing to investors via roadshow decks promising NFL-level popularity. Billionaire investors bought teams. In 2018, signed huge Twitch streaming deal, requiring in-game integrations like Twitch features, camera control, team skins, uniforms with technical challenges for memory footprint.
Plans derailed: no new world events, treading water, not focusing on Overwatch 2. Deals required make-goods; suggestion to refund money instead of burdening team. Too much focus on making money fast dragged people in.
Overwatch League brought great players and staff, but was a house of cards.
Shifted from quality esports ecosystem to money. Investors demanded input. In-person events failed due to global teams (e.g., London, Shanghai). Merch insufficient. Defaulted to live game revenue like $500 million prior year, pressuring team for sales.
Lost resources for live game events, heroes, maps. Believed in Overwatch 2 PvE version (not the shipped one); had hindsight on better design. Prioritized PvP maps as core audience; PvE expansion uncertain.
In 2016-2017, felt in control as game director with Ray Gresko as production director. Overwatch League and Overwatch 2 became albatrosses despite good intentions.
CFO demanded Overwatch hit specific 2020 recurring revenue (slipping to 2021), threatening 1,000 layoffs on Kaplan if failed. Biggest "fuck you" moment; surreal.
Meetings suggested scaling to Fortnite's 1,400 people and free-to-play. Believed would retire at Blizzard; this ended it. That CFO is gone.
Blizzard created hits via small incredible teams of weirdos taking risks, falling in love with craft. Not a machine; needs weirdos at helm. Businesspeople must get out of way.
Message to creators: love of craft lacks cutthroat ambition; stop handing golden goose to undeserving. World of Warcraft made without CFO; needed artists, engineers, designers, producers, audio.
Great finance people exist who respect and get out of way.
Leaving broke Kaplan; Blizzard is mecca, provided everything. Warm memories despite moments; painful mourning and grieving.
Started with Mike, Allen, Frank (gamers, programmers who made games). Protected developers, shielded from nonsense. Early COO Paul Sams protected. Joined at 95% developers/5% operations; left at 50/50 in 4,500-person company.
Commitment to excellence, high bar, passionate people like Chris Metzen, Sam Didier, Allen Adham. Infectious inspiration; wore Blizzard blue with esprit de corps, reverence for predecessors.
Best sessions. Early: got creative direction, e.g., Westfall zone ideas translated to gameplay.
Wrath of the Lich King: Asked about Northrend; drew maps, pulled Warcraft I/II manuals, old drawings. Listened, noted, photographed whiteboards, translated to design (zones, dungeons, sequence).
Countered Burning Crusade issue of everyone loading Hellfire Peninsula via Dark Portal; split Lich King into two starting zones for flow and choices (Sid Meier: games are series of interesting choices). Grizzly Hills example.
Metzen mixed pre-thought lore with on-spot creation; charismatic.
Kaplan's role: translate Metzen's visions into gameplay.
Burning Crusade Silvermoon City: Metzen wanted tallest tower in Azeroth for blood elves. Designers measured existing ( Blackrock Mountain, Karazhan, Stormwind Cathedral) to exceed; Kaplan clarified: make it really tall (vibe), took heat.
Diablo IV team crushing it; best modern Blizzard world. Understand demanding community. Rich experience beyond loot: art, seasons, masterful balance, new content at crazy pace, super fun. Great live team. Couch co-op excellent; one of greatest recent games.
"Never play to the gallery. Always remember that the reason that you initially started working was that there was something inside yourself that you felt that if you could manifest in some way, you would understand more about yourself and how you co-exist with the rest of society. I think it's terribly dangerous for an artist to fulfill other peoples' expectations. I think they generally produce their worst work when they do that. And the other thing I would say is that if you feel safe in the area that you're working in, you're not working in the right area. Always go a little further into the water than you feel you're capable of being in. Go a little bit out of your depth. And when you don't feel that your feet are quite touching the bottom, you're just about in the right place to do something exciting."
Took time off per wife's advice after 19 years, barely vacationing, consumed by Blizzard. Crushed by leaving. Pulled weeds (not garden), proficient. Played Call of Duty Black Ops Cold War, unlocked Dark Matter Ultra.
Late spring/early summer, started writing game ideas in Notepad. Terrifying: 19 years with greatest developers; used to delegate (e.g., Erin for world maps, Chris for story).
Jeff Kaplan rediscovered the pure joy of game making after leaving Blizzard. Surrounded by the best talent there, he worked solo on his own project, learning tools like Unreal, Adobe Illustrator, and Blender via YouTube. He realized he never wants to work for someone else again or have his creations taken away. He focuses purely on the craft: art, programming, design, audio—not the business side of the games industry.
Tim Ford, associate tech director on Overwatch, quit Blizzard in April 2021. Kaplan calls him his development soulmate. Ford skipped time off, arriving at Kaplan's house right after his exit interview to start programming their game. Kaplan pitched the idea and starting a company, birthing their studio.
Kaplan disappeared from public view to mourn leaving Blizzard and create independently, avoiding Reddit mentions or attention. Post-departure, 60 people reached out with investment offers amid a crazy VC and strategic funding market, especially from Chinese companies. He and Ford specified exact developer needs, timeline, and budget, optimizing for control rather than maximizing funds—no auctions.
Company: Kintsugiyama (deep personal meaning, explained later). Game: The Legend of California, an open-world action game set on a mythical island version of California during the 1800s gold rush. Ultra-realistic yet surreal alternate history: uninhabited island (not historically accurate), blending prospectors, cowboys, mines. Handcrafted world with familiar landmarks like Yosemite; voxels for generation. Dozens of points of interest (e.g., Dread Rock, inspired by Alcatraz) move by map seed. Map has four tiers of difficulty that shift per server seed (e.g., Mojave newbie on one server, endgame on another). Online multiplayer; gorgeous experience, music, mining. Integrates sound, lighting for immersion; best sunsets/sunrises in games, inspired by painter Albert Bierstadt's epic landscapes. Lighting artist Mike Marra replicates that feel.
Edgier tone: lonely, mysterious, larger-than-life, dangerous. Feel small until earning bigness; seek shelter at sunset, cozy up by ranch fireplace. Solitary yet beautiful, tense. Not Blizzard-style bright hero shooter or steampunk—authentic 1800s guns, no spaceships/aliens. Studio of 34 people defining Kintsugiyama games. Video games incorporate all prior art forms.
Quiet Steam launch for wishlists. Public-ish alpha in March. Then early access to evolve publicly—first time for Kaplan. More rough edges than Blizzard demos; exciting to show sausage-making.
Kintsugi: Japanese art of repairing broken pottery with golden lacquer, embracing scars as beautiful. Philosophy: nothing's perfect; beauty in imperfection, strength in broken things. Reflects Kaplan and Ford's Blizzard scars, game dev's pursuit of imperfection, personal aspiration.
Legend of Zelda: Breath of the Wild—greatest ever: thoughtful design, art, tech, Switch integration. Chopping trees, floating rivers; world as toy. Narrative, combat, itemization—all best-in-class. Lightness, openness unique to Nintendo, the Mecca.
- Others: Zork, Ultima, EverQuest (6K hours), Rust (5K hours), Red Dead Redemption 2 (shrine-worthy; brilliant craft, story, characters, Tarantino-level dialogue).
Rust inspiration: resetting worlds (monthly planned for Legend of California?); exciting resets for fresh adventures, equal starts (naked on beach). Not PvP-centric or full loot loss—Rust players love taking others' stuff, but Kaplan avoids that stress.
Zork: Nostalgic IBM PC memories with mom/brother, keeping lights on to avoid Grue.
Games: technology-driven art; invent camera each time. AI currently a hot mess—overconfident, low hit rate (e.g., wrong UI fixes in Unreal/UMG), weird/shitty outputs. Fun for non-artists (ChatGPT, Midjourney), but moral concerns: no one's creative work should be used by
No one's creative work should ever be used by AI without their permission. Voice actors and artists cannot have their work lifted without permission; that's immoral and equivalent to stealing.
Practical AI Applications for Small Studios
As someone running a small studio with 34 people, AI can help with points of tedium that one wouldn't hire for. Example: resizing 2,000 incorrectly sized images made in Photoshop. ChatGPT can resize them and zip the file in a minute. This saves hours of personal time without taking a job; no intern would be hired for it.
Human Creativity vs. AI Limitations
No matter how good AI gets, it will never draw a picture like Arnold Tsang or tell a story like Chris Metzen. The human spirit is irreplaceable.
Truly great creative minds and teams create something special that's hard to articulate. AI produces beautiful imagery, stories, and believable text, but lacks the human edge—perhaps the imperfections. AI currently feels like an interesting fever dream and a useful tool for mundane tasks.
Small studios are the future of gaming. Big studios acquire small studios for new IP and ideas, which then grow in. Compelling, new, innovative ideas will come from small studios.
Advice for Video Game Creators and Developers
To small teams wanting to create a truly special game: they already know how to do it. To video game developers in general: own the craft, own the art form, stop giving it to corporate jackals. You are the golden goose; keep your eggs.
Jeff, formerly from the Overwatch team, I have to say from the bottom of my heart, and I think I speak for millions of people, thank you for everything you've created in this world. Now that I've gotten the chance to see the new game, I can't tell you how excited I am to try it. Thank you for everything you've created. Thank you for everything you represent. Thank you for remaining and fighting for us as one of us.
Thank you, Lex.
Thanks for listening to this conversation with Jeff Kaplan. To support this podcast, check out sponsors in the description, with links to contact, ask questions, give feedback.
Don't bend. Don't water it down. Don't try to make it logical. Don't edit your own soul according to the fashion. Rather, follow your most intense obsessions mercilessly.
Keep Lex Fridman in your library
Save the videos and channels worth coming back to, and find them again in one place.
