1976: The Trucks Just Stopped Coming.
The lights were still on.No one knew how long it would last.Society didn't collapse overnight.It simply stopped showing up.
A leadership and social survival simulator set in April 1976.The trucks stopped arriving. Workers stopped showing up. Stores emptied slowly, then all at once. Guide a community through uncertainty, scarcity, and the collapse of the systems people once depended on.
Follow me on Facebook
1976: The Trucks just stopped coming.
1976 is a leadership and social survival simulator.Guide a community through uncertainty, scarcity, competing priorities, personal conflicts, difficult decisions, and the gradual collapse of the systems people once depended on.The challenge is not surviving a disaster.The challenge is leading people through one.
Why 1976?1976 is set in early 1976, before computers, smartphones, social media, GPS, and the modern internet.When systems begin to fail, people cannot simply search for answers. Communities must rely on local knowledge, relationships, leadership, and trust.The focus of 1976 is not technology. It is people.
1976 isn't a game about perfect information. It's a game about living in a world where everyone knows only part of the truth. People make decisions from observations, memories, rumors, trust, and experience. Sometimes they're right. Sometimes they're wrong. Sometimes they never learn the whole story.
In 1976, you don't select dialogue from a list.Simply say what you want to say—by typing or speaking naturally.Your words are interpreted in real time, and NPCs respond according to what they know, what they believe, who they trust, and what is happening around them.There are no dialogue trees to memorize or scripted conversations to exhaust. Every conversation develops naturally from the people involved, what they know, and the circumstances surrounding them.
A Living CommunityEvery NPC has a history that began before the player arrived.Families, occupations, friendships, rivalries, neighborhoods, and community ties existed long before Day 1.As the world changes, those relationships continue to evolve. Alliances form, trust is earned or lost, rumors spread, and communities adapt based on the choices made by both the player and the NPCs.
Built Around PeopleIn 1976, understanding someone is often more valuable than outlasting them.Leadership, cooperation, memory, trust, communication, and local knowledge become just as important as food, tools, and shelter.Sometimes the right conversation changes more than the right tool ever could.People don't respond because the script says they should. They respond because, from their perspective, it makes sense.People don't survive because they know everything. They survive because they learn from one another.Designed to Work Offline1976 is designed to run entirely on the player's computer.Core systems—including speech recognition, and NPC conversation—
are intended to operate locally without requiring a persistent internet connection or cloud-based AI services during gameplay.
Building the People of 19761976 has been in active development since May 2026. Much of that time has been spent on something players may never directly see—but should experience constantly: the cognition behind the people who inhabit the world.Rather than scripting an answer for every situation, we're building NPCs around their own observations, memories, practical knowledge, beliefs, concerns, relationships, and plans.The idea is simple to explain, even if it hasn't been simple to build:Give a character enough understanding to figure things out.An experienced character should approach a problem differently from someone who has only seen it done once. Someone may possess useful information but fail to think of it when they need it. A character may recognize that something is wrong without knowing how to fix it.Experience can matter without becoming a simple “skill level.” Characters can learn from what happens to them, apply related knowledge to unfamiliar situations, find another way when the expected solution isn't available, or simply fail because they don't understand enough yet.And people present another kind of problem.NPCs can pursue longer-term goals while dealing with other characters who have goals of their own. They may cooperate without completely trusting one another. They may reconsider a plan when circumstances change. They may choose what information to reveal, what to keep to themselves, and potentially even deceive someone when they believe doing so serves their interests.None of this is intended to make NPCs perfectly rational.People misunderstand things. They forget. They overlook connections. They make reasonable decisions from bad information. Two people can experience the same event and come away believing very different things about what happened.That's part of what we're trying to preserve.So... what AI is doing all of this?None.AI tools have been used extensively during the development of 1976. As an independent developer without a traditional development staff, they've provided access to expertise, review, programming assistance, troubleshooting, and the ability to challenge ideas from several different directions.But those tools aren't intended to become the intelligence running the NPCs.1976 is not being designed around ChatGPT, Claude, Grok, Gemini, or another online LLM deciding what a character should say or do.Instead, the cognition system is being built specifically for the game.That distinction matters. If a character doesn't know something, an outside AI shouldn't quietly provide the answer. What matters isn't everything the computer could possibly know.What matters is what this particular person knows.The mechanic, cashier, nurse, accountant, teenager, retiree, and player may all encounter the same problem with very different knowledge available to them.That's where their differences begin.What about dialogue?Conversation is intended to be part of that same simulation.What a character says can depend on what they believe happened, what they remember, what they're willing to disclose, who they're speaking to, and what they're currently trying to accomplish.A remembered conversation doesn't necessarily have to be a perfect transcript, either. Someone might remember the important part, forget a detail, retain their interpretation of what you meant—or remember something incorrectly.What you said, what they understood, and what they eventually remember don't necessarily have to be the same thing.Does the game depend on an online AI service?That's not the design.The goal is for 1976 to be a stand-alone game whose core cognition and dialogue systems operate as part of the game itself, rather than continuously querying an outside LLM.That means no AI account or AI subscription should be necessary to play. The goal is also that a persistent internet connection will not be required for gameplay. Internet access may still be used for ordinary purposes such as downloading the game, patches, updates, or future optional online features.But the people inhabiting 1976 shouldn't stop thinking because an AI server somewhere is unavailable.Then how do the NPCs do it?That's what we've spent the last several months building.We've broken cognition into interacting responsibilities: observation, attention, memory, learned knowledge, beliefs, concerns, planning, decision-making, dialogue, and behavior.More importantly, we've spent months trying to break those ideas.We've tested them against practical problems, incomplete knowledge, changing plans, conflicting information, social relationships, cooperation, deception, mistakes, improvisation, and situations where a character simply doesn't know enough to reach the correct answer.As those tests became increasingly different, something important began happening:We increasingly stopped needing new systems to explain new situations.Different problems began resolving through the same underlying cognition architecture.That doesn't prove we've solved the problem. Implementation and testing still have plenty of opportunities to prove us wrong.But it does mean the project has reached an important transition.Where development stands todayThe broad architectural discovery work behind the cognition system has now reached closure for this stage of development.The focus is shifting toward completing its implementation in Unreal Engine and verifying that the individual systems work together as intended.After that comes the part we've been working toward: putting these characters into controlled situations and seeing what they actually do.Not what we think they'll do.What they do.And when they surprise us, fail, misunderstand something, find an unexpected solution—or expose something in the architecture that doesn't work—we'll have our next problem to solve.AI helped us build it. An AI shouldn't be required to play it.And if an LLM isn't figuring everything out for the NPCs...How are they doing it?That's exactly the question 1976 is trying to answer.
Development Progress
Follow DevelopmentFor more frequent progress updates, screenshots, behind-the-scenes notes, and discussion, join
1976: The Trucks Just Stopped Coming on Facebook.The Facebook group is also a place to ask questions, share comments or concerns, and find out what
we're currentlyworking on as development moves toward Alpha and Beta testing.Join 1976: The Trucks Just Stopped Coming on Facebook
Future Project Exploration