AI at Google: our principles • Hacker News Board

Re: AI at Google: our principles

#181

Post by zaphar » Thu, Jun 07, 2018, 9:34 PM UTC

the-pigeon wrote: post #168

killjoywashere wrote: post #95

Earlier quoted context omitted.

You should read Daniel Kahneman's "Thinking, Fast and Slow". It's not possible to make all the right policy decisions that are right in hindsight and before a sentinal event occurs. Hindsight bias is always 20/20. Anyone making real decisions of consequence will eventually curse hindsight.

Sure. But Google has no intention of doing the right thing anymore than Microsoft or Disney does. These are corporations and their executives HAVE to do what they think will be best for the corporation. Not what they think is best for mankind. This is how for profit businesses currently work. And PR saying anything to the contrary is simply not true.

This is a gross generalization that people trot out as if it were unassailable but never back it up with any support.

Corporations are run by people with a complex set of motivations and constraints in which they make decisions. Some of them make decisions with intent to harm. Some make decisions with intent to help.

No one person is automatically turned into a ruthless amoral person just be being employed at a corporation.

Re: AI at Google: our principles

#182

Post by MarkMMullin » Thu, Jun 07, 2018, 9:36 PM UTC

hueving wrote: post #144

MarkMMullin wrote: post #85

Earlier quoted context omitted.

Yes. I can only refuse to play and hope the Russians et al come to their senses, the United Nations starts to work better, etc. If we go down that road, it will be the end of most of us, if not all of us. We really have no chance against a machine that's decided to shoot us. As a species we need to have zero tolerance for it - we won't be rendered extinct by snazzy terminators, we'll be killed by the 21st century ver…

This is a bit dramatic though because these robots can't reload themselves. Automated targeting for a single area is miles away from planning strategy, deploying battle groups, etc.

Not dramatic - the robots are not expected to reload - one robot one kill - simple as that- mass produced plastic, lots of robots, no intention whatsoever for reusability - the risk is that this means you dump thousands to millions of these stupid little drones, many make mistakes, and many others take naps and then wake up and do more damage when everyone thought it was all over. These are disposable killing machines, not Reapers. Furthermore, at todays prices, a cloud infrastructure and a million dollars can net you 10K lethal drones. There's no finesse, just brute force.

Re: AI at Google: our principles

#183

Post by jacobsenscott » Thu, Jun 07, 2018, 9:38 PM UTC

"Our AI is so powerful it needs special rules!" is pure marketing.

Re: AI at Google: our principles

#184

Post by themacguffinman » Thu, Jun 07, 2018, 9:41 PM UTC

EpicEng wrote: post #81

themacguffinman wrote: post #65

Earlier quoted context omitted.

I doubt that Google spelling out their moral stance is intended to convince you right away that they're all good now. It's a public standard that they're setting for themselves. If you think their actions don't match their words, you now have concrete terms and principles to critique and compare with. It's a benchmark to which employees and the public can hold them accountable.

>If you think their actions don't match their words, you now have concrete terms and principles to critique and compare with. What I think is that they will go forward with any project that has potential for good return if they don't think it will blow up in their faces, and that opinion is based on their past behavior.

I didn't realize they already have past behaviour of violating their own stated AI principles within the day of publishing those principles. /s

Doesn't sound like you're really that willing to give them the benefit of the doubt like you said.

Re: AI at Google: our principles

#185

Post by joshuamorton » Thu, Jun 07, 2018, 9:42 PM UTC

rmk wrote: post #174

joshuamorton wrote: post #152

Earlier quoted context omitted.

You appear to be suggesting that the government should be able to compel corporations (and thus their employees, ie. normal people) to work on military technology. That sounds antithetical to a lot of freedoms that we hold dear as Americans. >I would argue that the public should be especially wary of 'global' corporations such as these (Facebook is another one) that suddenly grow a conscience when it suits them. How…

"Google changes its policy to maintain its workforce" Google, an already known to be duplicitous company, changed its policy to maintain its workforce at the expense of US national interests . That's certainly something to be wary about, as a member of the public.

You didn't address the main point of my comment, so I'll ask again more directly: are you suggesting that the government create regulation that compels companies, (and thus individuals) to do work on military technology that it deems relevant to national interests?

If not, what exactly are you suggesting?

Re: AI at Google: our principles

#186

Post by commandlinefan » Thu, Jun 07, 2018, 9:42 PM UTC

chellam wrote: post #47

ISL wrote: post #4

The best way to lead is by example. Thank you, Googlers. The choice not to accept business is a hard one. I've recently turned away from precision-metrology work where I couldn't be certain of its intent; in every other way, it was precisely the sort of work I'd like to do, and the compensation was likely to be good. These stated principles are very much in line with those that I've chosen; a technology's primary pur…

Turning away will slow things down but AI for military applications will happen. Someone will fill the void.

> Someone will fill the void.

Especially since Google is publishing their results for every void-filler out there to review. Unless they plan to start hiding results that might have military applications?

Re: AI at Google: our principles

#187

Post by forapurpose » Thu, Jun 07, 2018, 9:46 PM UTC

At least they are starting the conversation. I'd be much more comfortable with principles of design and implementation in addition to outcomes. For example, transparency is essential. Also:

5. Incorporate privacy design principles.

We will incorporate our privacy principles in the development and use of our AI technologies. We will give opportunity for notice and consent, encourage architectures with privacy safeguards, and provide appropriate transparency and control over the use of data.

Why not "give people control over their privacy and over their information"? That's a commitment to an outcome. "Incorporate ... principles", "give opportunity", "encourage", and "appropriate transparency and control" are not commitments. Google seems to be hedging on privacy.

Re: AI at Google: our principles

#188

Post by ryandrake » Thu, Jun 07, 2018, 9:47 PM UTC

MarkMMullin wrote: post #177

ryandrake wrote: post #138

Earlier quoted context omitted.

That’s the problem with mutually assured destruction weapons (which AI bots most surely qualify as). If you accept as a given that at least one faction is building them, then the other factions’ only choices are: 1. Build your own and pray for a Cold War stalemate and 2. Don’t build your own and guarantee you’ll be destroyed.

MAD was based on the concept of the 'Nuclear Club' - the barrier to entry was the mass defect ratio, which you'd have to calculate on your own, and then you'd confront nation-state level expenses of turning that knowledge into practical engineering. It had an inherent cap on the number of possible players. Not saying it was nice, it was just manageable. The barriers to entry here are much lower. For smaller players,…

I guess my point is (and it looks like we agree): good luck getting all billion+ people on the planet with a credit card to agree that building a kill bot is a bad idea. All it takes is for one of those people to disagree, build them, and then we all must have them.

Re: AI at Google: our principles

#189

Post by mylons » Thu, Jun 07, 2018, 9:49 PM UTC

forapurpose wrote: post #187

At least they are starting the conversation. I'd be much more comfortable with principles of design and implementation in addition to outcomes. For example, transparency is essential. Also: 5. Incorporate privacy design principles. We will incorporate our privacy principles in the development and use of our AI technologies. We will give opportunity for notice and consent, encourage architectures with privacy safeguar…

they're only starting it because of the backlash from employees internally and more importantly the external backlash.

Re: AI at Google: our principles

#190

Post by rdiddly » Thu, Jun 07, 2018, 9:49 PM UTC

debt wrote: post #127

EpicEng wrote: post #55

So, I'm all for giving someone the benefit of the doubt if they have a change of heart upon reconsidering an issue, but this coming after the fact rings a bit hollow to me. I think the only principle at play here is that it became a PR issue. That's fine, but let's be honest about it. Early emails between Google execs framed this project only in terms of revenue and potential PR backlash. As far as we're aware, there…

I noticed there was one particularly phrase missing from Apple's WWDC this year: artificial intelligence. It wasn't mentioned one time. Never in the presentations, none of the talk titles had it included either. Machine Learning was mentioned and Apple also released many updates related to it, but never did they call it artificially intelligent, or use the phrase artificial intelligence. Not a single time. Go ahead a…

I credit their sense of discipline in using the more accurate term. They know they can only stand to lose down the road if they over-hype notions if "artificial intelligence" now when it really is a misnomer. And they're probably betting that's what will happen to their competition who more enthusiastically use the phrase for marketing.