Slashdot Mirror

You seem to be one of those people who is stridently confident in their ignorance.

Once upon a time nobody could imagine dish and clothes washing machines either.
We've already got industrial robots that can learn tasks by watching humans do them. Its not hard to imagine wealthy people buying something like that for their homes in 15 years, and those are the same people who currently hire house cleaners today. Or hotels.

Well, as far as we know...

Despite all talks of duplicate TNG/DS9 "rikers", it will be very difficult to destroy and replicate exactly someone using a "transporter" like device because of the No cloning theorm.

However, physics doesn't appear to preclude teleportation of some sort where the replication is dependent on the destruction. In fact people have been able to teleport quantum states of photons as far as a satellite in earth's orbit... Since photons are nominally the same except for their quantum state, that's basically teleportation.

Of course teleportation (in this quantum sense) requires pre-entangled objects to exist on both sides and merely theoretically allows...

1. an ingress device to destructively make classical measurements of quantum state of an original object in proximity of one of the pre-tangled objects,
2. take those classical measurements and communicate them to an egress device near the other pre-entangled object
3. an egress device is not longer prohibited from using the communicated measurements and the pre-entangled object to reconstruct the original quantum state in target object (which the act of which destroys the quantum state of the pre-entangled objects and overwrites the target object with the quantum state of the original object).

As you might imagine, this a whole lot easier, if the object to be teleported is a photon because we have relatively simple ways of generating pre-entangled photons, reasonably simply ingress devices to make measurements of quantum states of photons and actual egress devices to modify the quantum state of photons.

Beyond a photon, well, all this telportation stuff is currently very hard, but not yet proved theoretically impossible. ;^)

The Obama campaign invented the deep-dive into Facebook data for their 2008 and 2012 campaigns. They not only openly bragged about doing what CA did and far more, but somehow the media fawned all over him for it. Odd, that:

https://www.cnn.com/2012/11/07...

https://www.washingtonpost.com...

http://swampland.time.com/2012...

https://www.technologyreview.c...

Well as more and more jobs are taken by AI-driven automation, the bigger question is how many jobs will be left for humans?

Currently people work well in the 60's and even 70's and work 40 hours a week with a mere 2 weeks of vacation. OK so people retire in the 40's and 50's or change the work week to 32 hours, etc. Why not embrace the increase in productivity instead of fearing it or worse impede its inevitability?

This misses the point; companies are embracing automation (and "increased productivity") as a way of getting people off the payroll and the pension scheme. Who is going to pay people to retire in their 40s and 50s? Who is going to pay people an income they can live on for working part-time? Even today in the UK there are millions of people on "zero hours" contracts which do not guarantee them any work (and therefore any pay) in any given week/month. These people are technically not unemployed, but they have to rely on welfare to survive.

A future where we all somehow live lives of comfort and leisure while machines do all the work is a utopian fantasy.

And these:

https://www.cnn.com/2012/11/07...

https://www.technologyreview.c...

https://www.washingtonpost.com...

http://swampland.time.com/2012...

They BRAGGED about doing the same things (and worse) than what they're accusing Cambridge of doing.

Well as more and more jobs are taken by
AI-driven automation, the bigger question is how many jobs
will be left for humans?

Currently people work well in the 60's and even 70's and work 40 hours a week with a mere 2 weeks of vacation. OK so people retire in the 40's and 50's or change the work week to 32 hours, etc. Why not embrace the increase in productivity instead of fearing it or worse impede its inevitability?

Obama uses Big Data to great effect and is celebrated for it. Trump use it and OMG!

Perhaps in some circles. In others, the team you play for doesn't matter as much as what you stand for.

Well as more and more jobs are taken by AI-driven automation, the bigger question is how many jobs will be left for humans?

The idea that technology will magically create vast numbers of new jobs to compensate for the ones lost - the so-called luddite fallacy - doesn't work in a world where employers are deploying automation specifically to reduce their expensive human head count.

Obama uses Big Data to great effect and is celebrated for it.

Trump use it and OMG!

Just another in a long list of Hypocritical Controversies.

You mean the M7 motion-detection chip?

https://www.technologyreview.c...

But it'll be the white-collar middle class jobs that go this time, and that'll cause the chattering classes to squeal very loudly in ways they couldn't be bothered with when it was blue-collar robot automation that was affected.

Jobs like journalism - where Reuters already has the majority of its articles written by AI but it moving into news as well now:

https://www.technologyreview.c...

Other information systems will only go the same way. My biggest fear is that it'll be Google or the like providing these systems and creaming off all the data for themselves.

Google wanted to insert themselves into every transaction. Apple Pay was a more secure credit card.

Google didn't want to insert themselves; it was the only technically feasible option at the time that wasn't impossible to scale.

Yeah, right. They so much don't want to insert themselves that they're actively buying offline credit card transaction data from third parties.

On the contrary, I think Google salivates at the idea of inserting themselves into your wallet, and the deeper the better. They have a very strong motivation for collecting all your data and tracking all you buy - they can then use your purchases to show ad companies how efficient the ad buy is.

4. Talk to me about my investments.

So-called robo-advisors are already doing that in a limited way.

5. Diagnose my illness (without a doctor as the interface)

Again this is already happening.

6. Teach my kids.

It's already happening.

9. Rescue someone.

Well, for what it is worth Facebook apparently has an AI suicide prevention program. Rescuing someone does not necessarily require a physical act: mental problems are something that an AI might be able to help with.

Now it is certainly true that AI's roles in these areas are somewhat limited at the moment and there are somethings which it is hard to image AI being able to do within the foreseeable future. However, AI does not need to "do everything" to replace many jobs. If AI working in conjunction with a doctor lets that doctor diagnose 200 patients a day by identifying and dealing with the simple cases that reduces the need for doctors. Similarly if AI TA's let a professor teach 1,000 students effectively that reduces the need for teaching staff etc.

This is the way technology works: jobs change to do the work that technology cannot do with the result that a single human can do far more. Robots on assembly lines have not completely replaced all human workers but the work that humans on assembly lines do has changed to cover jobs that robots are not good at and to oversee the robots to fix things when they go wrong. In this way a handful of humans can run an assembly line that used to require a small army. This is not a bad thing: it lets us be far more productive with our time. However, care does need to be taken to ensure that it is possible for people to adapt to the changing jobs market and that things do not change so fast that it causes too much disruption for society to cope with.

Handled correctly changes like this give us more leisure time and a higher standard of living. Handled badly they can lead to civil unrest and worse.

Too bad you didn't look it up, because that isn't quite true. In the US, CO2 emission has been trending downwards for the last few years. While this is primarily due to natural gas use this is also due to the use of wind and solar power (which combine really well with natural gas since gas plants have very fast spin-up times) and more efficient cars. See https://www.technologyreview.com/s/601415/carbon-dioxide-emissions-keep-falling-in-the-us/. And two years in a row, global CO2 production declined http://physicstoday.scitation.org/do/10.1063/PT.5.1067/full/. So no, we are actually succeeding, not as fast as we need to, but the general trend is in the right direction. We can solve this, but if people keep falsely claiming that all we can do is mitigation then we're going to be in very bad shape. Moreover, the budgets for mitigation have been tiny in many locations.

Hmm... That is an interesting question, so I went to Google "gasoline fuel cell".

This is what I found:

Gasoline Fuel Cell Would Boost Electric Car Range

WSU researchers develop fuel cells for increased airplane efficiency

It looks like there are serious people working on it, but there's still a way to go.

You should choose your sources in life with greater care. Someone is using you.

• The Origins of the Net Neutrality Debate Telecommunications companies and their suppliers have been nursing dreams of tier pricing for years. July 8, 2006
• net neutrality - Etymology - Coined in 2003.

https://www.technologyreview.c...

Some people might survive global nuclear war. A planet with average temperature in the hundreds ÂC? Not so much.

But naaah! No big deal. I'm sure it's gonna be fine! Let's wait it out and see.

Base-load production isn't a subsidy or an externality, but if your wider point is about ignoring the deficiencies of one's own position, then I'd suggest people are usually aware of those deficiencies but tend to minimise them, or at least feel sure they're much less of an issue than the deficiencies of the alternatives.

My problem here is the false equivalency. You say that each side is equally bad for glossing over the weaknesses in their arguments, but when those issues differ by orders of magnitude it's not equivalent at all. While no solution is perfect, some answers can be dramatically better than others.

For your example, intermittent renewables are a solvable issue - with storage, with wide distribution and redundancy, with plenty of variation in the mix of sources. The solutions do add to their cost, but even so are cheap enough that they're still competitive. Compare that to fossil fuel emissions, which cost us hundreds of billions annually even before factoring in climate change, and still no effective solution exists. Billions more have been spent piloting "clean coal" plants, promising only to partially reduce emissions, but still without success. Unless you're aware of hidden costs to renewables that are anywhere near the vast costs of fossil fuels, I don't see how the two alternatives are remotely equivalent.

"Global Warming Will Kill Us All." Is probably a lot more likely than most climate scientists would like it to be - as unfortunately, the most accurate Climate Models tend to be the most pessimistic (I would love to be wrong in this)!

https://www.technologyreview.c...
"Global warming’s worst-case projections look increasingly likely, according to a new study that tested the predictive power of climate models against observations of how the atmosphere is actually behaving.

The paper, published on Wednesday in Nature, found that global temperatures could rise nearly 5 C by the end of the century under the the UN Intergovernmental Panel on Climate Change’s steepest prediction for greenhouse-gas concentrations. That’s 15 percent hotter than the previous estimate. "

Even the mildest predictions of Global Warming show increased threats to American security and economy - yet Trump claims that security and economy are more important than dealing with Global Warming:

http://www.latimes.com/world/e...
"The Trump administration will stay focused on economic growth and national security no matter the outcome of its climate change policy review, a U.S. official told delegates at a United Nations convention in Germany on Saturday.

http://www.syfy.com/syfywire/s...
"Navy Rear Admiral (retired) David Titley has stated very clearly that it's a threat to U.S. national security, and President Obama went as far, correctly, as to say that climate change denial is a threat to our security as well. Interestingly, current Secretary of Defense James Mattis — one of the very few people in Trump's administration who understands that climate change is real — has called it a threat to national security as well."

anyone who puts an omnidirectional mic in their home, tied to big-pig corporate, should expect no privacy.

Note: cell phones and even laptop mics aren't very omnidirectional. You can also use a cell or laptop with a movable mic cover.

OTOH, the whole point of a smart speaker is to listen and snoop.

Agreed. Still, there might be a market for a less mistrusted version of these devices. The problem is I presume you would need to dedicate a reasonably powerful computer to it, since you would be keeping all processing local, and then only interacting with the rest of the Internet when a command is confirmed. Perhaps it could be something available via apt-get or similar. Then the microphone/speaker component is just connected to the real computer over wifi. Of course if you can get enough local processing to determine when the device is being talked to, then that bit can be done locally, and the actual message sent back to the more powerful system, which could then do a fast wake, process, and then return of the result.

A house could then have multiple visible nodes, but only one processor node. Multiple houses that share family may be able to link them over a VPN, with appropriate limits. The processing node would maintain the VPN.

This kind of thing would be the only way I'd have such a system in my house. At that point you may be able to include interface modules, such as something that controls televisions and such, but those are simple enough to be less concerning.

For instance, do you really need to have an action like, "Turn on lights" have any potential traffic outside your house? Of course for that you probably need a limited network over house wiring, that is available to the processing node, but not the rest of the Internet.

Out of curiosity what open source projects like this exist? I found this with a 10 second search by a company called MyCroft. link

It's not certain that this reduces pollution, as agriculture has significant run off of fertilizer and pesticides and there are wastes from chemical refinement into biodesiel. As for carbon footprint, even that is not clear that soybean biodiesel is a net benefit.

Of course, I would help a lot with credibility if the molten-salt people do their homework correctly, before making outstanding claims.

https://www.technologyreview.c...

This technology is being developed to replace things like high-paid ghost-writers of pop music and screen writers in Hollywood. With the current state of the entertainment industry, these jobs should be easy to automate.

Andrew Reagan and others at the Computational Story Lab at the University of Vermont in Burlington have used sentiment analysis to map the emotional arcs of over 1,700 stories and then used data-mining techniques to reveal the most common arcs. “We find a set of six core trajectories which form the building blocks of complex narratives,” they say.

• The Shapes of Stories (Computational Story Lab, University of Vermont)
• Six Basic Emotional Arcs of Storytelling (MIT Technology review)

* * * * *

• The Five Key Turning Points Of All Successful Movie Scripts (Movie Outline)

* * * * *

• Movie Narrative Charts (XKCD)

Considering oil is a non-renewable source.

STOP RIGHT THERE. We can turn CO2 into gasoline. Given that MANY applications (such as airplanes) simply aren't viable without the extreme energy density of fossil fuels relative to batteries, trying to eliminate all fossil fuel use seems like folly. Given that we can actually make gasoline and kerosene from CO2, why isn't there more effort put into that?

You do know that you can spend bitcoin in other places than China...
You do know that you can use exchanges in other places than China...
You do know that you can mine bitcoin anywhere you can get electricity... including China...

Did you read the links...Using bitcoin isn't banned, just exchanging Yuan to bitcoin and ICO's are banned. Some exchanges are still and will be still open to convert any bitcoins back into Yuan for the foreseeable future. They are just cracking down on sending Yuan outside China by exchanging them to bitcoin first.

you are clueless

The latest restrictions are more draconian, with cryptocurrency exchanges now shut down. But once again, workarounds have emerged. Some people have turned to online and offline peer-to-peer trading. People can also buy and sell digital currencies on the encrypted messaging app Telegram, which is blocked in China but can be accessed by virtual private networks (VPNs) that get around the Great Firewall. People who already own coins can just go online and trade them on an exchange that is based overseas. There was even some trading on WeChat, China’s massively popular but heavily monitored messaging app.After all, China did not ban Bitcoin itself, nor did it explicitly prohibit peer-to-peer trading. And importantly, China hasn’t banned the mining of bitcoins, in which people have their computers race to solve difficult mathematical problems in exchange for coin rewards.

the only rational choice is to think about how to manage it.

OK, so how does one "manage" a technology that nobody really understands (and the only realistic shot at changing that currently on the horizon seems to be "let the algorithm explain why it does what it does")?

Indeed. That's a very, very hard problem. Roughly half of the book is about it. It's an even harder problem than you think it is, because not only do we need good solutions, we need good solutions that can be implemented even though we don't have the ability to force everyone researching AI to implement them. So we either need solutions that can be implemented over the objections of some groups, or we need solutions that everyone actually wants to implement.

My core point, though, is that "let's just not build AI" is not an option, because there's no way to prevent people from doing it.

the only rational choice is to think about how to manage it.

OK, so how does one "manage" a technology that nobody really understands (and the only realistic shot at changing that currently on the horizon seems to be "let the algorithm explain why it does what it does")?

I get that theory in the abstract, but when nobody understands why their own AI algorithms decide to do what they do, what exactly do you propose we do to have a prayer of black-boxing algorithms of outlaw AI, or whatever else you mean by "keep ahead of it"?

Think about it... When the 1% deletes anything written by anyone else, then everything will be written by the 1%.

Yep. That's the problem. There are a small number of editors who believe that they personally own the articles they wrote, and will revert any changes made by anybody else. And, since they do this deletion a lot, they are very good with the Wikipedia bureaucracy and know exactly how far they can go without getting counted as "edit warring"-- and how to entice novice editors into breaking one of Wikipedia's invisible rules and getting banned.

The article says : "As detailed in a 2013 feature in the MIT Technology Review, the decline of active editors with more than 10 edits under their belt has been attributed to the increasingly bureaucratic nature of the editing process. The semi-automation and stricter editing process was initially launched as a way to combat vandalism on Wikipedia pages. Although the new protocols did result in a decrease in vandalism, it also resulted in a steep drop off of new editors that stayed 2 months after their first edit."

No. It's not the semi-automation, it's the bureaucracy being used by the "deletionists" who don't want you-- if you fail to follow obscure rules when responding to the asshole who deletes the stuff you just wrote, you will be banned.

It's always fun to be called out by ignorant fuckwits who want to push an agenda, but the facts are not on their side:
https://www.technologyreview.c...
"The inflection point has already been reached in the West, and by 2021 solar will be cheaper than coal in China."

Yes, fracking has historically been the reason coal is going away, but renewable sources are now adding to that.
"Those milestones will surely lead to greater adoption of clean energy. And the report predicts that of the $10.2 trillion expected to be invested into power generation between now and 2040, 72 percent will be channeled into renewables."

So crawl back under your bridge and STFU.

CCS is just an excuse for coal plants to continue to operate and to try and promote coal as "clean". It only captures a small percentage of the CO2 and it is expensive and maintenance intensive and doesn't scale. For example, the Kemper County plant was a failure with trying to gasify coal to reduce the CO2 emitted then capture the remaining CO2 after spending billions over budget. Coal is dying and Trump's pulling the clean power initiative won't save it and will actually make things worse for those who live in coal country. The CPI had a program in place to help train people for other lines of work. Now coal use will decline but without the program to help the workers displaced by this. Coal has no future. All of the cheap coal is gone and fracking has made natural gas far cheaper (despite the problems like earthquakes due to fracking).

Germany pays subsidies for wind and solar only to export the electricity at low prices so it's not a complete loss. Not only low prices but negative, as in paying people to take it so the grid remains stable. They are bankrupting themselves.
http://fortune.com/2017/03/14/...
https://www.technologyreview.c...
http://euanmearns.com/getting-...
http://www.windpowermonthly.co...

Denmark imports electricity at 30 euro/MWh and exports at 20 euro/MWh. Germany does better with imports at 30 euro/MWh and exports at 27 euro/MWh. If Germany keeps shutting down reliable nuclear and replacing it with unreliable wind and solar the net export is likely to disappear, the price difference is most definitely going to spread, and this will cost Germany money. Perhaps Germany will remain a net exporter of electricity but they will have to pay their neighbors to take it.

These "environmentalists" like to talk about things being sustainable. This is not sustainable.

I feel like I'm reading a response from 50 years ago, when we were all going to have flying cars and live in glass boxes and only have to work 20-hour weeks because computers would do everything for us.

Moore's law is dead, I wouldn't count on it for future planning. Any space savings is going to come from custom ASICs, which means you're going to be even more at the mercy of the manufacturer for spare parts (and you'll never be able to ride in a "classic" self-driving car, as they'll all be dead from bit rot). Speaking of which, who's going to service these self-driving beasts? I know my dealership doesn't have any techs with degrees in telematics or sensor networks, and I suspect a lot of them would quit if they had to start learning how to use logic probes, signal injectors and expert systems.

I think most people are well aware that Puerto Rico is in no position to pay for an island wide solar power. Consider that an island of 3.4 million people recently declared bankruptcy on $70 billion of debt. This was in the news and people are well aware just how destitute they are, therefore they realize just how absurd it is for the island to pay for solar power.

There is also the matter of looking at their power needs. The people their can't afford the cost of solar energy as they already pay higher electrical costs than 49 out of 50 states. They pay these costs when making far less money than people in the United States - and that's before the hurricane wiped out far too many jobs.

https://www.eia.gov/state/anal...

Their problem isn't their power plants, they are largely intact. Their problem is the power lines, lines that would still have to be rebuilt even if they did use solar.

https://www.theverge.com/2017/...

Even MIT has debunked the ludicrous idea of Puerto Rico rebuilding with solar. In short they need about 20 billion megawatt hours per year. Tesla's south Australia facility will produce 129 million megawatt hours. The two are an order of magnitude apart in scale.

https://www.technologyreview.c...

To quote MIT on the matter "And given estimates that restoring the grid could take up to six months (not including Teslaâ(TM)s involvement), one is left wondering if the cost, complexity, and longevity issues donâ(TM)t make the suggestions rather more bluster than substance."

To say that Tesla could or would actually build out something that would meet the needs of Puerto Rico is absurd for anyone other than Tesla to say.

In the real world, resources are always finite.

How many years did it take the auto industry to be shamed by Ralph Nader into providing safety features for their customers?

Suggesting that these cars will intentionally prioritize avoiding mechanical damage over human life is absurd.

Depends on the business model. Not every business model will prioritize human life. Based on the business model is how these self-driving cars will be programmed.

https://www.technologyreview.com/s/542626/why-self-driving-cars-must-be-programmed-to-kill/

About 40% are how to switch to Google as default.

60% are from Microsoft's AI, Tay trying to search how to meet sexy alt-right single men.

Tay keeps re-submitting because bing thinks she wants to buy packs of American Singles. White cheese, of course!

Our best bet to get a closer look any sooner is to use our Sun as a gravitational lens. It is still a challenge, because we would need to put a telescope at the correct side of the Sun at about 550 AU, far beyond the orbit of Pluto, but it is much closer to our technological reach than actual interstellar probes. NASA is thinking about this project: https://www.technologyreview.c...

While this batch of bacterially infected mosquitos will (hopefully) induce a short term (months? years?) drop in the population, a "technical" fix for permanently solving this problem worldwide now exists.

It is called "Gene Drive" if you haven't heard of it and what it utilizes is the powerful genetic editing technique called CRISPR. Basically you alter the genes of a male mosquitos so it carries the CRISPR gene package so that all of its children are male. Then, the included CRISPR package in the genome alters all of THEIR genes so that all of their children are male (and carries the package forward).

The population because more and more male "dominated" until there are no females left. Then poof!, after one last generation, they're gone.

While the deliberate (we're doing it all the time accidentally) elimination of a species is obviously something that shouldn't be done lightly, since THIS particular species carries Malaria, Dengue Fever, West Nile virus and Zika it would seem to b a prime target. Bill Gates, after his foundation spent several hundred million dollars trying to eradicate Malaria says he's all for it because he thinks it may be the only way to wipe out some of these terrible scourges (millions of dead children). It appears as if New Zealand will try this technique to get rid of an invasive species, a mammal(!) introduced by European settlers; the mouse. (If successful they plan to continue doing this to many other invasive species).

https://www.technologyreview.c...

Of course it would be the height of irony if a mosquito managed to transfer the CRISPR gene package (from itself or a mouse) to its main host, thus getting rid of the most invasive species in Earth's history: US

"Bias" in these articles (and the studies they talk about) mean that the algorithms' predictions are wrong in opposite directions for people of different groups. The ProPublica article compares the AI predictions with the actual recidivism of ex-prisoners, and looks at the rates of "false" positives and "false" negatives for blacks and whites. The cases in the MIT article also compare predictions with results.

"Bias" in these articles (and the studies they talk about) mean that the algorithms' predictions are wrong in opposite directions for people of different groups. The ProPublica article compares the AI predictions with the actual recidivism of ex-prisoners, and looks at the rates of "false" positives and "false" negatives for blacks and whites. The cases in the MIT article also compare predictions with results.

"Further, the fact that more people of a particular race are prosecuted is not a reflection of bias in the data, rather a bias in the prosecution."

In this case, "persecuted" was more accurate.

Data is Data. It cannot exhibit a bias.

I can only surmise that you're not an experimental scientist. Data has bias all the time.
In physics (my field) the bias usually has no social consequence-- astronomical statistics, for example, are biased toward bright stars (since they're much easier to see than faint ones, and hence overrepresented in the data set). In social "sciences," however, the bias very often does have social consequences. SAT scores from children whose parents spend tens of thousands of dollars on SAT Prep courses, for example-- surprise!-- score better on SAT exams than ones who don't. The data shows a correlation of SAT score with parental income. Is this real? Better correct for the SAT-prep course effect before making a conclusion.

Data is biased. All the time. Be ready for it.

...Plus, being from the Guardian, I am skeptical that they didn't twist the data some to obtain their desired outcome, which ironically touches on the subject of this story.

Huh? MIT Tecnology Review and Propublica were the source. The link in the summary was this: https://www.axios.com/algorith... which linked here: https://www.propublica.org/art... and here MIT Technology Review

There are good reasons to be against GMO, mostly because the commercial part of it or the side effects that "change".

So because of some commercial uses bother you, means the technology is bad? That's like saying we should ban desktop computers because Microsoft makes questionable business practices. Besides, many well known GMO patents have expired, and farmers have been using GMO plants royalty free for the last two years to the exclusion of non-GMO because they know the GMO ones to be superior:

https://www.technologyreview.c...

Example: You do realize GMO plants do not contain pollen?

This is just downright laughable. By far the most prominent GMO product is canola, which without a doubt produces pollen. Read myth #2 here:

http://www.npr.org/sections/th...

The whole anti-GMO movement relies on mis-truths, and wouldn't exist at all if nobody ever made up any of the bullshit you just bought in to.

Top google hits, try it next time.

"This is one of the difficulties of using the term artificial intelligence: it's just so tricky to define. In fact, it's axiomatic within the industry that as soon as machines have conquered a task that previously only humans could do - whether that's playing chess or recognizing faces - then it's no longer considered to be a mark of intelligence. As computer scientists Larry Tesler put it: "Intelligence is whatever machines haven't done yet." And even with tasks computers can beat, they aren't doing it by replicating human intelligence."
https://www.theverge.com/2016/...

"Artificial Intelligence is the broader concept of machines being able to carry out tasks in a way that we would consider "smart".
Machine Learning is a current application of AI based around the idea that we should really just be able to give machines access to data and let them learn for themselves."
https://www.forbes.com/sites/b...

"Machine learning is a particular approach to artificial intelligence. It is true that it is proving to me the most successful approach to AI. But, I disagree with Monica Anderson's answer: it is NOT the "only" approach.
For example, you'd be surprised to hear that some of the self-driving cars that currently describing themselves as using AI, use very little machine learning and are mostly using rule-based systems."
https://www.quora.com/What-are...

About the problems of marrying concepts whose relationships are not well understood:
https://www.technologyreview.c...

There have been plenty of real advances in the last few years, not just speed improvements. For me, the most impressive thing was Generative Adversarial Networks in 2014, but there have been plenty of advances. The most recent article I read was on Relational Reasoning https://www.technologyreview.c... .

Here as some more recent advances
Turing Learning - https://www.sheffield.ac.uk/ne...
Evolution Strategies - https://www.technologyreview.c...
Bayesian Program Synthesis - https://techxplore.com/news/20...
Gaussian Processes - https://www.wired.com/2017/02/...
AI Passes Standard Intelligence Test - https://phys.org/news/2017-01-...
Semi-Supervised Learning For Handwriting Recognition - https://phys.org/news/2016-12-...
Lipreading - https://www.technologyreview.c...
One-Shot Learning - https://www.technologyreview.c...
Differentiable Neural Computer - http://spectrum.ieee.org/tech-...
Bayesian Program Learning - http://www.ctvnews.ca/sci-tech...

There have been plenty of real advances in the last few years, not just speed improvements. For me, the most impressive thing was Generative Adversarial Networks in 2014, but there have been plenty of advances. The most recent article I read was on Relational Reasoning https://www.technologyreview.c... .

Here as some more recent advances
Turing Learning - https://www.sheffield.ac.uk/ne...
Evolution Strategies - https://www.technologyreview.c...
Bayesian Program Synthesis - https://techxplore.com/news/20...
Gaussian Processes - https://www.wired.com/2017/02/...
AI Passes Standard Intelligence Test - https://phys.org/news/2017-01-...
Semi-Supervised Learning For Handwriting Recognition - https://phys.org/news/2016-12-...
Lipreading - https://www.technologyreview.c...
One-Shot Learning - https://www.technologyreview.c...
Differentiable Neural Computer - http://spectrum.ieee.org/tech-...
Bayesian Program Learning - http://www.ctvnews.ca/sci-tech...

There have been plenty of real advances in the last few years, not just speed improvements. For me, the most impressive thing was Generative Adversarial Networks in 2014, but there have been plenty of advances. The most recent article I read was on Relational Reasoning https://www.technologyreview.c... .

Here as some more recent advances
Turing Learning - https://www.sheffield.ac.uk/ne...
Evolution Strategies - https://www.technologyreview.c...
Bayesian Program Synthesis - https://techxplore.com/news/20...
Gaussian Processes - https://www.wired.com/2017/02/...
AI Passes Standard Intelligence Test - https://phys.org/news/2017-01-...
Semi-Supervised Learning For Handwriting Recognition - https://phys.org/news/2016-12-...
Lipreading - https://www.technologyreview.c...
One-Shot Learning - https://www.technologyreview.c...
Differentiable Neural Computer - http://spectrum.ieee.org/tech-...
Bayesian Program Learning - http://www.ctvnews.ca/sci-tech...

There have been plenty of real advances in the last few years, not just speed improvements. For me, the most impressive thing was Generative Adversarial Networks in 2014, but there have been plenty of advances. The most recent article I read was on Relational Reasoning https://www.technologyreview.c... .

Here as some more recent advances
Turing Learning - https://www.sheffield.ac.uk/ne...
Evolution Strategies - https://www.technologyreview.c...
Bayesian Program Synthesis - https://techxplore.com/news/20...
Gaussian Processes - https://www.wired.com/2017/02/...
AI Passes Standard Intelligence Test - https://phys.org/news/2017-01-...
Semi-Supervised Learning For Handwriting Recognition - https://phys.org/news/2016-12-...
Lipreading - https://www.technologyreview.c...
One-Shot Learning - https://www.technologyreview.c...
Differentiable Neural Computer - http://spectrum.ieee.org/tech-...
Bayesian Program Learning - http://www.ctvnews.ca/sci-tech...

Close the N.S.A.'s Back Doors. (New York Times, Sept. 21, 2013)

NSA's own Hardware Backdoors May Still Be a "Problem from Hell". (MIT Technology Review, Oct. 8, 2013)

This 'Demonically Clever' Backdoor Hides In a Tiny Slice of a Computer Chip. (Wired.com, June 1, 2016)

Expert Says NSA Have Backdoors Built Into Intel And AMD Processors. (Eteknix, 2014)

When spyware is detected, that particular vulnerability is fixed:

Red alert! Intel patches remote execution hole that's been hidden in chips since 2010. (The Register, May 1, 2017)

Intel Active Management Technology, Intel Small Business Technology, and Intel Standard Manageability Escalation of Privilege (Intel Corporation, May 5, 2017 ) Quote: "Severity rating: Critical"

I came here to say the same thing. I remember reading about Obama's data operations in an article like this one: https://www.technologyreview.c...

The DNC moped the floor with the RNC in terms of data analytics and social media use during Obama's campaign. Did they suddenly dismantle that when Hilary ran? I have a hard time believing that they didn't even make progress on it. At the same time, I was skeptical that the RNC had improved their data analytics over the last 4 years, and while I'm sure they did, I have to imagine the comparative advantage went to the DNC.

Hilary's cognitive dissonance is really astounding. I hope this pisses off the DNC and causes them to go for someone else in 2020.

I'm not a fan of cap-and-trade for carbon (or anything else)-- see, for example, https://www.technologyreview.c...

But when you say "our solution of carbon credits is a global scam used to..."
--wait, WHAT "solution of carbon credits"? This system doesn't exist.