Now, no one does.
I can see the allure of moving away from a legacy cobol system. But an AI rewrite doesn't actually solve any of the issues with having a legacy cobol codebase. You just have a new system no one knows or understands.
I got a chuckle over this, but in my experience working on legacy systems, this is the case already. Using AI to translate/explain the code brings extra understanding.
I think COBOL systems would be plenty maintainable if the owners of the systems invested in maintaining them. If companies just let people retire, attempt to hire replacements with unremarkable salaries, then maintenance is a problem.
If they paid salaries that reflected how critical this work is, I don't think it would be that big a problem to find and keep good developers.
We have critical economic infrastructure being treated as annoyance and a cost centre.
IMHO every large software system should be developed on the basis of clear and concise functional specifications so everyone can understand what the software does and how it does so at a higher functional level.
well, we just need to learn the rewrite then ???
Not true, AI does. It's AI job security...
I do wonder how an ambitious AI would handle the maths. COBOL is renowed for its fixed no rounding maths, utterly rigid and defined in the code, Java on the other hand is not (e.g., 0.1 + 0.2 becomes 0.30000000000000004), unless they are extensivly using BigDecimal, then its already a broken migration.
I've seen projects proclaim in the past to migrate COBOL systems to Java, and fail expicly when the solution was a dtaawarehouse system and a seperate Java system using the legacy backend to do all the fancy marketing reports or whatever other department was pushing for changes they wanted without a grasp of what it truly entailed and blinded by some sales pitch.
It's not just the code that needs migrating, its the data as well, then you want solid robust infrastrucure, which is why IBM still sell those mainframes to run all that well battle tested COBOL code.
Is it impossible to migrate to Java or any othger language - no, anything can be done, but the level of devil in the details and not just software, the data and a robust system to run it upon. It is very easy for even the best to overlook a gotcha.
We have all used an AI where it has gone down one path and you then point out something and it corrects itself, without the knowledge and a load of old grey beads(or a team) to act as devil advocate and try to pull the output apart, YOU can end up with something that works for everyones satisfaction, but is hiding a few gotchas down the line. AI coding needs a full debate team with one puishing it and one countering it almost I'd say.
If it works, why change it is always a good approach and why we mostly have in certain systems a COBOL backend with APIs or datawarehousing to expose for fancy java or other language added value/functionality.
I can only image generations of directors pushing cutting edge to replace legacy systems, and learning that cutting edge can cut you. My sympathy for the IT departments who have to deal with that as it has always been a bain of IT departments who are then forced to justify why it's fooling in a constructive way and explain it to non IT people in upper managment why the case.
As for migrating COBOL to anything, I'd not do it myself, id suggest a rewrite/redesign of a new system from scratch and then run both in parallel for a long time and make sure they actually do the same job and results. Even hand migrating back in the day, would produce a lot of messy code that refactoring would make sence and yet, not what the client would pay for as that would be a rewrite.
Prompt the core architecutre well suited for it.
Let it write tests.
Let it vibe migrate from left to write.
Now you at least have a chance of a team being able to work on that code base and actually clean it up and make it good.
Cobold is hard, hard legacy with bad tooling.
From the paper:
> The COBOL source is passed through an internal deterministic Migrator to produce a generated Java target.
Also, humans are not deterministic either. Give the same COBOL -> Java translation to multiple developers and each will come up with a different solution. Heck, even the same developer will produce a different output for the same task, depending on the day of the week.
Just set the sampling temperature to zero and remove any unintended non-determinism during the parallel computation of the token probability distribution. The problem is solved? Of course, not. Non-determinism has little to do with LLMs' mistakes.
Most compilers are also not deterministic, at least by default, but they don't usually make mistakes. Determinism isn't an important quality here. And if it were, AI can be deterministic, it just isn't normally for much the same reason compilers typically aren't (hint: performance).
I find it bizarre that I keep reading this here. Just one of those things that keeps getting blindly repeated without receiving any thought?
Correction: AI is not deterministic, the only realistic low-error solution is not a more complex use of non-deterministic AI, but deterministic transpilation.
The problem is that this results in COBOL-in-Java which runs correctly but it is a nightmare to maintain.
At least if tescoverage is good, but well... That's something llms can also be used for
There are hundreds of billions of lines of cobol in production.
The IRS alone has approx 160 cobol programs, averaging 230kloc each.
Their pitch is like, "we take your COBOL code, and all the real-world data you can give us. Then we model your COBOL program as a graph, where each node has inputs and outputs. Then we use AI to port each node, making sure that it has the same (input => output) mapping for all the test data you gave us."
This is the secret sauce. You'll be responsible for any defects if you didn't prepare your test data to act like a massive unit test. Wonder how hard is to prepare the ideal test data vs writing the unit tests themselves in COBOL and verifying their translation. I guess if you throw enough data at it, the effort is minimal while the coverage becomes good enough to fix any bugs manually.
https://www.pega.com/insights/resources/break-free-legacy-ma...
https://www.ibm.com/products/watsonx-code-assistant-z
https://global.fujitsu/en-global/pr/news/2026/03/30-01
https://www.rocketsoftware.com/en-us/insights/ai-powered-cob...
Maybe.
Still not something you want to risk your business on.
Related to this, I created a tool for situations where you want to compare COBOL code with Java code. It includes a preprocessing step where IMS etc. calls are converted to mocks that return JSON (from file), and also use JSON for input/output, and GnuCOBOL to run the program. More a proof-of-concept than production, but here is a link if somebody finds it helpful.
CICS is the big issue but AFAIK there were attempts. Everything else, including sort and VSAM, has various options provided usually by COBOL compiler vendors.
COBOL code:
PROCESS-DATA.
ADD 1 TO COUNTER.
DISPLAY COUNTER.
IF COUNTER < 10 GO TO PROCESS-DATA.
Java translation: while (counter < 10) {
counter++;
System.out.println(counter);
}
If "counter" is 10 on entry, the COBOL code prints 11 while the Java code prints nothing. So not only keeping old bugs, but apparently introducing new ones too!I wrote COBOL code for a few years at a job when I was a teenager. What makes legacy COBOL code difficult IMO is it can sometimes be very hard to maintain a mental execution state when examining the code, for several reasons:
1. all variables are global, aka, WORKING-STORAGE. You list all the variables used in the program and they are accessible to the entire program.
2. programs are divided into paragraphs. Control normally flows sequentially top to bottom through paragraphs, one executing after another. Except that the PERFORM statement can drastically alter this normal flow control, and you can't tell by looking at a paragraph how it will be executed. To do that, you have to look at all PERFORM statements that mention this paragraph or any paragraph physically before it, because in COBOL you can say PERFORM PARA1 THROUGH PARA27. If PARA13 is physically between PARA1 and PARA27, it's potentially going to get executed.
3. In true legacy COBOL, before structured COBOL was a thing (circa 1985), the main control flow statement in addition to PERFORM was GOTO. Lots of flag setting, and lots of GOTOs. So in the previous example, you can't tell if PARA13 is going to get executed because any prior statement might be a GOTO PARA14, skipping execution of PARA13. But even worse, you are still under the influence of the PERFORM THRU, so after PARA27 is executed, control returns to the statement following the PERFORM THRU, wherever that was. But if you GOTO PARA27, without being under a PERFORM THRU, then PARA27 is executed followed by the next sequential paragraph. Trying to figure this out statically by looking at the program can be very difficult, especially considering PERFORMs that are nested at runtime but may not be anywhere near each other in a code listing.
As specified, please find 99997 correct parts and the 3 defects (do not use)
I recently threw in (want to check those intelligence metrics!) some real production code into a non-agentic system (just to get a feel how things perform without a custom harness) and results where ... interesting. The particular program uses some preprocessor no LLM we have access to (newest was GPT 5.5) has any clue about - so they confabulate what it could do (Gemini 2.5 didn't even notice there was a preprocessor...). This is expected of course but it somehow seems the problem of this technology that unless you feed it masses of data or mechanically break up the tasks in rote subunits, it just doesn't do anything sensible still...
I don't remember their name, hopefully they stayed in business. since probably convincing financial firms that using Rust is better than Java is a mammoth task.