Voice of the DBA

Tag: AI

The IT Jobs AI Can’t Do

This was a provocative title: 6 ITOps Skills That Will Never Be Automated. In a time when AI use is growing quickly and many people fear for their jobs, it’s nice to see someone writing about areas that AI will struggle to handle.

Ironically, lots of these articles are written by writers without much technical expertise and who are more likely than others to struggle in the age of AI. If a computer can mimic styles, less writers will be employed. That might be fitting with so many journalists hyping technologies without really understanding them, doing so in a way that creates stress among actual IT professionals.

The article lists 6 jobs: policy config, incident response, complex scaling, app feedback, ITOps tool deployments, and end-user support. Of these, I’m not sure that app feedback and end-user support are good-paying jobs that many people want, but I know there certainly are people working in these areas who depend on those jobs.

The others are often jobs that few people do. I do think that these are complex jobs where a computer can’t really take in enough information to do a great job here, but I do think that one smart worker could learn to guide an AI that does a lot of the busy work. Someone could specify a policy from an AI and then ask the AI to tweak it as holes or gaps are identified. I think over time a lot of quick incident response items could be handled by an AI, albeit with a human guiding it. Again, that reduces the need for humans in these roles.

To me, that’s where the AI revolution becomes scary. AI is a lever to get a lot done. Fewer humans can be used to accomplish tasks with the aid of AIs, which reduces a lot of the labor needed. After all, we know that in any area, there are lots of beginners and advanced beginners doing work. If we can use an AI assistant to help one or two humans, we might remove the need for a dozen others.

Maybe that’s what a future 10x engineer looks like. Someone that’s way more productive than many others because they’ve leveraged the computer (in the form of an AI) to get a tremendous amount of work done.

I do think there are likely jobs that AIs can’t easily do, but I also think that a lot of the work in these areas might get copy/pasted between organizations. Policies, scaling choices, and more are things that one human might learn from others and mimic those efforts. With better judgment than an AI, but not as much as if a true expert were coming up with the best solution for a particular situation.

Unfortunately, most of us don’t come up with optimal solutions, and our organizations run fine with sub-optimal code/policies/decisions/whatever. Will companies get by with very few highly paid people or a few more low ones and accept mediocre software?

That already happens far too often already. I suspect like most trends, we’ll see quite a few organizations move in this direction. However, I think the human ingenuity will win out and as a few companies realize that the creativity of humans can better target their goals than an AI that merely summarizes other work, the pendulum will swing back.

Unfortunately, a lot of people will get caught in bad situations as this happens. Work on your skills, technical and soft, and be sure that you are in the best position you can be in as the world embraces and evolves with AI technology.

Steve Jones

Listen to the podcast at Libsyn, Stitcher, Spotify, or iTunes.

August 30, 2023
ADS Copilot Experiments with XML

I sent some code to a customer recently to help them decrypt some stored procedures. I sent a quick and dirty set of code, noting at the bottom that the results were in XML and needed to be extracted.

The customer wrote back that my code produced XML and the procedure code had to be copied and pasted into SSMS to create a decrypted procedure.

Before I spent time on this, I decided that I should play with Copilot here. This is the perfect place and it’s what I’d hope at some point if I sent that code, the other person would use some sort of AI to help them fix things.

Maybe not, but here’s what happened.

This is part of a series of experiments with the ChatGPT and other AI systems. Lots of Copilot lately.

Asking for Help

I could look up the syntax for working with XML, but what about Copilot? Let’s see what happened.

An interesting suggestion. One problem: when I run this, the result isn’t great.

If I cast this as XML, or declare it, things work. At least, they don’t produce errors. But they don’t do what I wanted.

I wasn’t sure what to do, so I opened the completions panel for Copilot and saw other suggestions.

Let’s try these.

Suggestion 1 looks good. If I change my declaration to be XML, this works (or cast things).

Some points for Copilot here, and I’d hope a junior would get to change the declaration or ask how to convert the variable to XML.

Suggestion two doesn’t work.

Suggestion three looks like a copy paste from a forum somewhere. However, the code works. I don’t know how this gets into the suggestions, but I am interested to know what’s happening here.

Four and fix aren’t great. They look like repeats. However, six accounts for my declaration.

Does the code work? It does.

I accepted solution 6 to see what happens. The code was added to the query window.

A Repeat

I’ve given feedback, so let’s try again. I went back to the prompt and got the same suggestion again, however when I opened the panel, I saw different items.

One doesn’t make any sense and doesn’t work. Two and three are nonsense. Four is really interesting, but not useful.

I don’t think Copilot learned anything, and I’m not even sure it is all that well trained. I don’t know if it doesn’t recognize it’s in a database editor or what. The type of language for the ADS file is SQL, so I don’t know what X# and Java are added.

I’m out of patience today. I know how to do this, so I’ll just write the code.

August 23, 2023
A Real World Clash with AIs

A group of people are protesting more cars in San Francisco by disabling self-driving cars with a cone on their hood. It is likely that the cone disables some sensors and won’t let the cars move. For truly autonomous cars, this could be an issue if people could prevent them from working with small hacks. I could imagine a strip of duct tape or some other smaller obstacle that might also work and be less noticeable to anyone looking at the vehicle.

Ignoring the humor here, or the more-cars-in-cities debate, this is an interesting place where computing systems can fall down. People do find ways to hack sensors or other ways in which our software gets data from the real world, which can force behaviors we didn’t expect. We might allow for these, or we might not. In this case, I don’t know that the developers could do anything different, other than perhaps detect and alert the owners that the vehicle can’t move until someone clears the obstacle. They could add more sensors, but I suspect hackers would just block those.

The bigger issue is that as we depend on computing devices that interact with and need to work in the real world, we’ll have more problems like this. We might see simple hacks, or even everyday events that can interfere with the machines completing their tasks. If we get very dependent on these AIs, then these hacks become more than just vandalism. Imagine getting in this vehicle in an emergency to go to the hospital, and it won’t move because a sensor is blocked? That could be more than inconvenient.

We certainly have similar issues with people in the real world. People flatten tires (how does an autonomous car change a tire), they padlock doors and gates, and they might do any number of things that interfere with others’ ability to live their lives. Humans can often adapt to these issues with small behavior changes, but what about computing devices? Will a new class of jobs come about to correct issues like this? Will we see “fix-it” robots that autonomously move around and remove cones from hoods or change tires?

It’s hard to even think about all the crazy permutations of how autonomous computing devices can evolve and the types of interactions we might have. I’m not even sure if I think that the way we are approaching AI and autonomous systems is good. We’re often applying similar frameworks to them that we do for other humans, which I’m not sure if the best way to approach these decisions.

The future is looking very strange, and it’s coming quickly. I’m not even sure what to think about issues like this.

Steve Jones

Listen to the podcast at Libsyn, Stitcher, Spotify, or iTunes.

August 21, 2023
Using AI with Data Tasks

The hot new technology of the year is AI. Between ChatGPT, Copilot, and generative AI, it seems that this is invading the world of computing at an incredible rate. Whether this becomes really useful and valuable or not is something that we will seem over time. There have been plenty of trends in this area in the past that haven’t become as ubiquitous as the hype would lead you to believe.

I’ve done some light experiments with AI on my blog. To date, I haven’t found this to be that useful, other than a few cases where I basically used an AI to search the web for me. Rather than read a bunch of SSC or Stack Overflow results, the AI summarized things.

Somewhat.

I definitely had to test and verify the code more than I feel I’ve done with code posted in a forum. Of course, I do less experimenting because the AI results were a little more targeted to what I needed, rather than my cobbling knowledge and partial solutions together. I’m also not sure which I prefer.

I would like to use AI for data work, and there is an article that talks about some of the ways that we’ve used AI in the past. Data profiling has made sense, and I can see value here. For data security, I’m not sure how helpful AIs have been. I’ve looked at some products, and I don’t know that I think any of them do a great job of identifying data. They do make it easy for whoever is assigned the task by doing some of the work, but they aren’t a panacea. They make mistakes, just like humans do.

I do think data observation and looking for anomalies is a place where AI can really shine, but that’s not the data work that many of us do. It matters, but for most of us, this isn’t something we deal with.

The future of AI was more interesting. The idea of data homogenization, taking data from different sources, and fitting it to a data model is interesting. Of course, the AI can’t make too many mistakes, or the time correcting might overwhelm the time saved. I think we see that now with humans who we ask to ETL data. If they aren’t good at it, or make lots of mistakes, those of us overseeing them might just do the entire job ourselves.

I know that AIs are still new and immature, and while there is a lot of potential, they feel like junior staffers now, needing more handholding and micromanaging than I like to do. Perhaps they will change our careers and the way we work, but I don’t know how quickly, or even how deeply. Already I find lots of companies putting restrictions on what their employees can do with AI, which makes me think this might be more a targeted, niche technology more than a general, use-it-everywhere solution.

Steve Jones

August 19, 2023