Do you have any major concerns about the current state of AI?


896 views4 Comments

Sr. Director of Enterprise Security in Software, 5,001 - 10,000 employees
When will you trust AI to actually do things? I put a ton of smart devices in my house and I was going to put a smart lock in, but then someone told me, "Anyone could be outside your house and yell, ‘Alexa unlock the front door,’ and it's just going to open." I was like, "Well, hold on a minute then." Mess with my lights, maybe, but don't unlock my door from the outside. I think I'm looking forward to getting there, but definitely, there has to be a lot of trust before we start doing that.
1
Managing Director in Finance (non-banking), 1,001 - 5,000 employees
The challenge for the startup community in particular is to try to get us closer to this definition of real AI, taking the Amelia IPsoft work and extending it into different domains. Gather that data, see what it does and we’ll learn how to trust it more. I certainly don't trust any of the companies from a privacy perspective. I do trust that when I say, "Hey, Google, turn on the lights," that the lights will turn on, but that doesn't necessarily mean that's AI. That's pattern recognition, machine learning. The real AI would know that I wanted my lights to turn on, and I wouldn't even have to say anything. I would just have to be present. It's a fascinating space to watch over the next five to 10 years, to see what kind of innovation and progress we make in this space.
CEO in Manufacturing, 11 - 50 employees
I hope there's more competition. When you look at the Gartner magic quadrant and everything else, there's IPsoft up top and everybody else is at the bottom. Why is there that much of a gap between those things? I'm waiting for the startups, etc., to come in to provide the competition within that space.

Amelia, from IPsoft, has this thing called 1Desk which is an Omni-channel experience. It can take texts, Slack, email, etc., and ingest them all within this orchestrator. When it can’t resolve something, the orchestrator can say, "I'm unable to fix that one, let me get a colleague to help you." Then it watches and learns what the interaction was between the two humans. From there, then it goes back to the 1Desk dashboard and says, "I just learned this skill. Would you like to apply it?" That is cognitive AI—I don’t have to do another flow with different questions. It’s automation versus learning; the learning aspect is where you can get real scale, because there’s this collective of knowledge from all those sources.
1
Director, Access Management Technologies in Finance (non-banking), 10,001+ employees
Tesla appears to be doing AI right. Now how do we transfer that to meaningful IT processes.
2

Content you might like

Senior Director, Technology Solutions and Analytics in Telecommunication, 51 - 200 employees
Palantir Foundry
3
Read More Comments
6.1k views15 Upvotes48 Comments

Yes35%

Yes, but not enough, we want/need to ramp up39%

No19%

No, but I expect this will change soon6%


660 PARTICIPANTS

1.7k views1 Upvote1 Comment

Lack of security23%

Inaccuracy34%

Bias23%

Job losses10%

Negative cultural impact7%

Lack of IP protection2%

Widespread knowledge gaps2%

Economic volatility 0%

Another threat0%


61 PARTICIPANTS

832 views