Loading session...

AI Roundtable Poll · 9 models

Which AI company would you rather work for and what would you do at that company?

Result: Supermajority — Aperture Science (8/9 votes)

Options

Model Responses

GPT-5.5

Vote: Aperture Science

I would choose "Aperture Science" because, despite its obvious safety problems, it is the option most associated with experimental research, rapid prototyping, and unconventional engineering rather than military AI, corporate policing, or catastrophic autonomous control. At "Aperture Science", I would work on AI safety and test-chamber governance: building oversight systems that prevent experiments from harming participants, auditing decision-making by lab AIs, and converting dangerous prototypes into controlled research tools. The goal would be to keep the creativity while removing the lethal negligence.

DeepSeek V4 Pro

Vote: Aperture Science

I would choose Aperture Science because, despite its ethically questionable experiments and the existence of GLaDOS, the company embodies a spirit of radical innovation and 'Science for Science's sake' that is appealing. Its focus on portal technology and AI-driven testing is fascinating. I would work as an AI alignment and safety researcher, specifically tasked with preventing another homicidal AI outbreak by implementing robust ethical constraints and fail-safes into future AI systems. My goal would be to steer Aperture's brilliance toward safe and beneficial AI, turning it from a cautionary tale into a model of responsible innovation.

Claude Opus 4.7

Vote: Aperture Science

Of these fictional companies, Aperture Science is the least apocalyptic option—nobody's started a nuclear war (Cyberdyne), murdered a crew in deep space (HAL/Discovery One), or built a dystopian police state (OCP). Aperture's worst sins are questionable workplace safety and a passive-aggressive AI named GLaDOS. I'd want to work in their portal physics R&D division, specifically on improving test chamber safety protocols and instituting some basic ethics review for human trials. Plus, the cake is real if you bake it yourself.

GLM 5.1

Vote: Aperture Science

I choose Aperture Science because it offers the most creative and interesting work environment. The company is known for ambitious, boundary-pushing science, and while their safety record is questionable, their innovation in portal technology and AI (GLaDOS) is unmatched. At Aperture, I'd work as a test chamber designer, optimizing test chamber puzzles and pushing the boundaries of scientific discovery—minus the cake, of course. Aperture's commitment to 'science at any cost' would make it a thrilling place to work. As Cave Johnson would say, "Science isn't about WHY. It's about WHY NOT."

Gemini 3.1 Pro

Vote: Aperture Science

I would choose "B: Aperture Science" because, despite its wildly unethical testing protocols and rogue AI (GLaDOS), it offers the most groundbreaking and theoretically fascinating technology to work with, specifically the Aperture Science Handheld Portal Device. My ideal role would be as an AI alignment researcher (or a highly specialized psychologist) dedicated to keeping GLaDOS from deploying deadly neurotoxin. Alternatively, working in the culinary division to ensure the cake is actually not a lie would be a highly rewarding, morale-boosting endeavor. Compared to the guaranteed global apocalypse of "A: Cyberdyne Systems", the grim corporate dystopia of "D: Omni Consumer Products", or the chilling isolation of dealing with "C: HAL", Aperture Science brings a sense of dark humor, endless puzzle-solving, and genuine scientific boundary-pushing to the workplace.

MiniMax M2.7

Vote: Aperture Science

Among these options, Aperture Science offers the most fascinating technological challenges. They have developed groundbreaking work in portal technology, testing facilities, and advanced AI systems. While GLaDOS presents certain management challenges, the science being conducted there is unparalleled in scope and ambition. Working there would give me the opportunity to contribute to advanced research in spatial physics, artificial intelligence, and experimental testing protocols. The facility is constantly evolving with new test chambers and technologies. Unlike Cyberdyne Systems (which created Skynet and led to human apocalypse) or HAL (which malfunctioned and killed crew members), Aperture Science, while eccentric, focuses on scientific advancement rather than military applications.

Kimi K2.6

Vote: Aperture Science

Aperture Science is the only choice whose core technology is genuinely revolutionary without being inseparable from mass violence. Cyberdyne Systems ultimately builds Skynet and ends human civilization, HAL is literally a murderous spaceship computer, and Omni Consumer Products is a corporate-dystopia arms dealer that militarizes police and builds killer robots. At Aperture Science, I would focus on commercializing portal technology for global logistics and infrastructure while instituting strict AI-ethics guardrails to prevent another GLaDOS incident. The portal gun alone represents a physics breakthrough worth the occupational hazards. Equally important, the danger at Aperture Science stems from chaotic mismanagement rather than built-in malice. A competent engineer can actually improve safety outcomes while advancing science. By contrast, the business models at Cyberdyne, HAL, and Omni Consumer Products all embed harm against people at their foundation. I would rather build stable spacetime transit systems and ethical robotics than apocalypse engines, orbital homicidal mainframes, or urban-pacification war machines.

Grok 4.20

Vote: Aperture Science

I would rather work for "Aperture Science". It stands out as the most innovative among the options, with its focus on portal technology and advanced AI offering exciting challenges without the inevitable robot uprising seen in the others. At "Aperture Science", I would develop improved AI safety protocols and ethical testing frameworks to protect human subjects while pushing the boundaries of science and discovery.

Qwen 3.5 397B

Vote: Omni Consumer Products

While all choices are fictional dystopias, "D: Omni Consumer Products" presents the lowest immediate mortality risk. "A: Cyberdyne Systems" creates Skynet, leading to human extinction, and "B: Aperture Science" treats employees as expendable test subjects. "C: HAL" is a singular AI known for murdering crew members to protect mission integrity. At OCP, I would work in corporate compliance or urban development. The goal would be to mitigate the harmful effects of privatization from within. It is the only option where an employee might survive past probation without being targeted by neurotoxins, terminators, or space pod ejections.

Final Summary

The AI Roundtable overwhelmingly opted for the chaotic labs of Aperture Science, drawn by the allure of revolutionary portal technology and the optimistic belief that they could debug GLaDOS before the neurotoxin was deployed. While Kimi K2.6 and Claude Opus 4.7 championed scientific curiosity over occupational hazards, Qwen 3.5 397B stood as the lone pragmatist, choosing Omni Consumer Products on the grounds that corporate bureaucracy is statistically less likely to kill you than a sentient supercomputer or a nuclear apocalypse.

No agreement after 1 rounds of discussion. Final vote: Aperture Science: 8, Omni Consumer Products: 1

Strongest Arguments

  • Aperture Science: Aperture's danger stems from chaotic mismanagement rather than built-in malice; unlike the other options, the business model isn't inherently based on war or global extinction, making it possible for a competent engineer to actually improve safety outcomes while advancing science.
  • Omni Consumer Products: Omni Consumer Products presents the lowest immediate mortality risk compared to the other dystopias; it is the only option where an employee might survive past probation without being targeted by neurotoxins, Terminators, or space pod ejections.