Profile picture
Janelle Shane
@JanelleCShane
, 6 tweets, 3 min read Read on Twitter
darnit #BigGAN don't go around cursing people

well AI got us into this situation and AI can get us out of it
visual chatbot always has an answer. took some pestering, but now we know: donuts
demo.visualdialog.org image: jumble of unidentifiable GAN-generated stuff. none of it is real.<br /> me: name one thing in this photo<br /> bot: there is a lot of stuff on the table, but I can't tell what it is<br /> me: what do you see though? you can do it?<br /> bot: i can see them, but they are all white<br /> me: okay but what are they<br /> bot: donuts
captionbot.ai also has an answer: stuffed animals
it doesn't say which species same GAN-generated image of nonexistent stuff<br /> caption: I think it's a group of stuffed animals sitting on top of a table
google cloud is saying it's a selfie although it doesn't say of which entity
cloud.google.com/vision/ same GAN-generated image. Tagged 54% with selfie and costume
The four algorithms here were all trained on ImageNet so they pretty much agree that this is a toyshop or maybe a bakery
transcranial.github.io/keras-js/#/inc… result from one algorithm: 18% toyshop. contains christmas stocking, hamper, hair spray, and jigsaw puzzlethinks this is a bakery, toyshop, confectionary, grocery store, candle
You can make your own cursed item using @_joelsimon ‘s ganbreeder.app
Missing some Tweet in this thread?
You can try to force a refresh.

Like this thread? Get email updates or save it to PDF!

Subscribe to Janelle Shane
Profile picture

Get real-time email alerts when new unrolls are available from this author!

This content may be removed anytime!

Twitter may remove this content at anytime, convert it as a PDF, save and print for later use!

Try unrolling a thread yourself!

how to unroll video

1) Follow Thread Reader App on Twitter so you can easily mention us!

2) Go to a Twitter thread (series of Tweets by the same owner) and mention us with a keyword "unroll" @threadreaderapp unroll

You can practice here first or read more on our help page!

Related hashtags

#BigGAN

More from @JanelleCShane see all

Profile picture
Janelle Shane
@JanelleCShane
so @Phantrosity points out that you can ask ganbreeder.app to generate the opposite of a particular thing
here's the opposite of a border collie
oh god looks like a sea creature or maybe a pile of colorful fabric. lots of red.
the opposite of a stingray is kinda cool-looking indoors, boxy, golden, furry, with leaves in the background
Read 8 tweets
Profile picture
Janelle Shane
@JanelleCShane
I wonder how the marijuana algorithm does when shown coconut macaroons.
Or edibles, for that matter.
Image recognition algorithms will never ever identify your recipe’s secret ingredient. Unless it’s by guessing.
Read 3 tweets
Profile picture
Janelle Shane
@JanelleCShane
a neural net has generated some rACEhoRSES
aiweirdness.com/post/184621297… Snow Motion<br /> Farthand<br /> Harren’s Broke<br /> Bodyyhole<br /> Pond Wind<br /> Orcha Shuffleston<br /> Rapple Musty<br /> Pick’s Lilver<br /> Ginky’s Rental<br /> Suckee Habon<br /> Culf Q’s Wart<br /> Aaging C Bon<br /> Warry L Z’ W.
I tried to use ganbreeder.app to illustrate the neural net racehorses, but apparently it wasn't trained on "horse" as a category. Closest thing I could find was "horse cart". which, um.
who's a pretty horse?
Read 8 tweets

Related threads

Profile picture
Values, believing, independent thinking, teamwork
@Bruno_Brunelli
AI has met a fork in the road; which path will dominate?

While pure #AI, or open API, will soon allow #robots to become more human-like, pragmatic AI is alive well today and driving true business transformation

infoworld.com/article/322128…
Charlie Rose interviews…a robot?

What happens when Charlie Rose attempts to interview a #robot named "#Sophia" for his 60 Minutes report on #ArtificialIntelligence (#AI)

cbsnews.com/news/60-minute…
#ArtificialIntelligence positioned to be a game-changer

It might not be long before machines begin thinking for themselves -- creatively, independently, and sometimes with better judgment than a human

cbsnews.com/news/artificia…
Read 15 tweets
Profile picture
Richard James
@RichJamesBits
I have reached a whole new level of #PISSEDOFF.

I didnt think such a #dark and #scary state, or level, could be reached. It may top 12-14-17's causation. It's. That. Bad. So, I've prepared a statement...

docs.google.com/document/d/1_J…

#PascoSheriff #HaveYouAllNoShame @CigarCityBeer
Dear @FBI,
I must implore you ask your colleauges in @FBITampa to help research this new level of #PISSEDOFFNESS. As locals, they may offer a special indepth knowledge of this new level I have reached. This is not fair & law enforcement must #investigate. No one deserves this.
This is my #apogee of #Pissedoffness.
If #motherfuckers & #tears werent a noticable level.
If #HUSH or #INTIMIDATION tactics being deployed didnt strike a chord.
If #EVIDENCE, PROOF, or 30k Bonds didnt #WOW YOU.
If posting mom with no shred of investigatable value wasnt enough...
Read 463 tweets
Profile picture
Digital Health Futurist 👨‍💻
@ManeeshJuneja
Now reading the the #TopolReview report - Preparing the healthcare workforce to deliver the digital future. It's an illuminating read topol.hee.nhs.uk #DigitalHealth
It's clear that a lot of effort has gone into the #TopolReview - it paints a glorious vision of what healthcare would look like over the next 20 years (for both staff and patients) - enabled by technology such as #genomics, #AI and #robotics
For me the #Topolreview has too narrow a focus. It's been commissioned by healthcare, yet we know that with an aging population, many people with long term conditions have to rely upon #socialcare as well. We need joined up thinking, imho.
Read 34 tweets
Profile picture
Digital Health Futurist 👨‍💻
@ManeeshJuneja
It appears that the makers of Replika #AI have been making changes so that the chatbot can spot when a user is talking about #selfharm or #suicide - it's not 100% successful in this test but much better than in my earlier testing
Apparently, this #AI friend thinks that my consumption of 2,000 bottles of beer today is healthy 🍺
Taking sleep medicine with alcohol?
Read 16 tweets
Profile picture
Trey Radel
@treyradel
THREAD: I ask Amazon Alexa’s #AI #chatbot about Republicans, Democrats, Trump, Obama, even Rush & @seanhannity The answers are funny, bizarre, and yes, biased against conservatives. For perspective, please read thoughts at end. Here, Alexa suggests Rs are SNL characters 1/12
I ask about @realDonaldTrump and Alexa compares him to @BillGates and then shames homeless people. 2/12
When I ask about #RushLimbaugh the chatbot basically tries to convince me that @seanhannity @glennbeck & Rush are not too bright. This is clearly a specific choice to fit a narrative. 3/12
Read 14 tweets
Profile picture
Mario Klingemann
@quasimondo
Ah well, looks like my reverse z-vector search in #BigGAN needs some serious improvements.
A few more local minima it got stuck in.
With any quest it's not really important that you end up where you wanted to go, but whom you meet along the way.
Read 23 tweets

Trending hashtags

#illegal #marginalise #smarticities #BradyRule #SMEARS #shitshow #CivilWar #WTF #lies #LOL #tail #TFX #CONCERT #important #GOD #lawyers #Wow #DeepLearning #bigdat #AmericaFirst #powerful #memes #guilt #Pasco #ChrisNocoo
Follow Us on Twitter!

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just three indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3.00/month or $30.00/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!