Täällä seurataan AI-ripuligeneraattoreiden maailmanvalloitusta

[pigra senlaborulo]

Why some celebrities are embracing Artificial Intelligence deepfakes

Singaporean actress, model and former radio DJ Jamie Yeo has no problem with being deepfaked. In fact, she signed up for it.

"It's a bit like that Black Mirror episode with Salma Hayek," Ms Yeo jokes.

She was speaking to the BBC the day after the release of the new series of Charlie Brooker's Netflix show. In the first episode, actress Salma Hayek, playing a fictionalised version of herself, signs away her image to a production company.

The deal allows it to use an artificial intelligence or AI-generated deepfake version of the Hollywood A-lister to "star" in their new TV drama. What she says and does in the show is controlled by the computer.

The consequences for Ms Hayek - without spoiling the story - are not good.

Concerns about the impact of AI are partly behind the first Hollywood actors' strike in more than four decades, bringing the US movie and television business to a halt.

It comes after Screen Actors Guild (SAG-AFTRA) failed to reach an agreement in the US for better protections against the misuse of AI for its members.

The actors' union has warned that "artificial intelligence poses an existential threat to creative professions" as it prepared to dig in over the issue.

However, Ms Yeo is not worried. She is one of a growing number of celebrities embracing AI-generated advertising.

The new technology is being met with a mixture of excitement and trepidation.

Ms Yeo has just agreed a deal with financial technology firm Hugosave, which allows it to use a digitally manipulated likeness of her to sell their content.

The process is fairly simple. She spends a couple of hours in front of a green screen to capture her face and movements, then a couple more hours in a recording studio to capture her voice.

An AI programme then synchronises the images with the audio to create a digital alter-ego capable of saying practically anything. The results are uncanny.

"I do understand the concern, but this technology is here to stay," she says. "So even if you don't embrace it because you're scared, there will be other people who will embrace it."

Some already have. As part of his deal with PepsiCo, superstar footballer Lionel Messi allowed it to use a deepfake version of himself to advertise Lay's crisps.

Not only can online users create personalised video messages from "Lionel Messi", they can get him to say it in English, Spanish, Portuguese and Turkish.

Fellow football superstar David Beckham and Hollywood legend Bruce Willis have also dabbled with deepfake technology - though, unlike Ms Yeo, they have so far stopped short of signing away full image rights.

"I think deepfakes will just become part of normal practice in the advertising industry over the next few years," says Dr Kirk Plangger, a marketing expert at King's College London.

"It opens the door to all kinds of creative options. They're able to micro-target consumers and are often extremely persuasive."

The efficiency of the process also makes it attractive from a commercial point of view.

"You're not doing that much work for the money you're charging," Ms Yeo says.

"It's also good for the client on a budget because they get so much more content than from a normal shoot. So it works for everyone."

The client - in this case Singapore-based Hugosave - agrees.

"Having this technology available means we can literally produce hundreds of videos in a matter of days. Compare that to the months, if not years, that we'd need if we were filming the content in the traditional way," says Braham Djidjelli, Hugosave's co-founder and chief product officer.

"We're able to leverage the benefit of AI while also retaining the human touch of a trusted local face - in this case Jamie's."

But, as analysts such as Dr Plangger point out, there is a "dark side" to the technology.

"It's not something we can put back into the box," he says. "The advertising industry needs to wake up to the risks as well as the possibilities of artificial intelligence. It means stepping back, as a society, and thinking about what is the proper or ethical use of this technology."

One of the things Dr Plangger is referring to is a looming "crisis of trust", where consumers cannot tell between what is real or fake. This is something already being exploited by vested interests online and can range from synthetically manipulated pornography to misinformation to political messaging.

But there are also more practical risks for the talent voluntarily signing up to be deepfaked. Currently, there are no clear laws relating to AI to ensure your image is properly protected.

For example, what happens if a brand uses your digital avatar to endorse a product that may damage your image or your alter-ego makes a joke in poor taste?

"We are in uncharted territory when it comes to AI and deepfake technology," says Tng Sheng Rong, an intellectual property lawyer at Rajah and Tann in Singapore.

"Many, many issues can arise. Who owns the intellectual property? Who do you go to for legal recourse? The truth of the matter is that the existing laws don't provide a robust enough regulatory framework to guard against these issues."

That may be the final hurdle for advertisers before they start buying rights to digital versions of Hollywood A-listers, for example.

At this early stage, Ms Yeo says she is very mindful of the risks, but her decision was informed mainly by trust - both in Hugosave and in the way business is done in Singapore.

But ultimately, she says, it is about staying ahead of the curve.

"If you want to still be in the game, then you've got to learn how to be in it. Because if you don't, you should probably just retire."

https://www.bbc.com/news/business-65995089

[pigra senlaborulo]

Current Artificial Intelligence (AI) models are "quite stupid", Facebook-owner Meta's president of global affairs Nick Clegg said as he played down the risks of the technology.

The former UK deputy prime minister said the "hype has somewhat run ahead of the technology".

Current models were "far short" of warnings where AI develops autonomy and thinks for itself, he said.

"In many ways, they're quite stupid," he told the BBC's Today Programme.

He was speaking after Meta said it its large language model known as Llama 2, would be free for everyone to use, known as open source.

Large Language Models - the platforms which power chatbots like ChatGPT - are basically joining dots in enormous datasets of text, and guessing the next word in a sequence, he said. He added that the existential threat warnings issued by some AI experts relate to systems which don't yet exist.

Meta's decision to make Llama 2 widely available for commercial businesses and researchers to use has divided the tech community. In some ways its hand has already been forced - the first generation, Llama, was leaked online within a week of its launch.

Open-source is a well-trodden path in this sector - opening up your product for others to use gives you an enormous amount of free user testing data, identifying bugs, problems and improvements along the way.

But the risk here is that this is a very powerful tool, whatever Sir Nick might say.

We know previous iterations of chatbots have been manipulated to spout hate speech, generate false information and give harmful instructions. Are the guardrails strong enough for Llama 2 to prevent itself from being misused out in the wild, and what will Meta do if it is?

Another interesting thing to note is Meta's decision to partner with Microsoft on this - Llama 2 will be available and usable via Microsoft platforms such as Azure - as Microsoft has also invested billions of dollars in ChatGPT creator OpenAI.

This is a giant with its sights firmly set on AI, and the deep pockets to buy its way in with the key players. The risk is that the AI pool soon becomes full of a few very large fish - and is that healthy for competition in this still fairly young industry?
Open-source

Llama 2 marks a partnership between Microsoft and Meta.

Unlike Llama 2, GPT-4 and fellow rival Palm - Google's LLM which powers the Bard chatbot - are not free to use for commercial or research purposes.

It comes a week after US comedian Sarah Silverman announced she is suing both OpenAI and Meta, alleging that her copyright has been infringed in the training of the firms' AI systems.

Dame Wendy Hall, Regents Professor of Computer Science at the University of Southampton, told the BBC that allowing AI to be open-sourced raised concerns around legislation.

"My worry about open-source is that is how we regulate them," she said.

"Can the industry be trusted to self-regulate, or will they work with the governments to regulate? It's a bit like giving people a template to build a nuclear bomb."

Sir Nick said her comments were "hyperbole", and clarified that Meta's open-sourced system could not generate images, let alone "build a bio weapon".

But he "strongly agreed" that AI needed to be regulated.

"Models are being open-sourced all the time already," he said.

"So it's not really whether open sourcing of these large language models is going to take place, the question is how can you do it as responsibly and safely as possible.

"I feel I can assert without any fear of contradiction, that the LLMs that we are open-sourcing are safer than any of the other AI LLMs which have been open-sourced."

https://www.bbc.com/news/technology-66238004

[pigra senlaborulo]

Meta on julkaissut generatiivisen kielimallinsa LLaMAn toisen version, LLaMA 2:n. Toisin kuin ensimmäinen versio on toinen versio julkaistu avoimena lähdekoodina, ja se on vapaasti hyödynnettävissä paitsi ei-kaupalliseen myös kaupalliseen käyttöön.

Meta julkisti asian tiistaina blogissaan. Samalla se ja Microsoft ilmoittivat uudesta kumppanuudestaan tekoälykehitykseen liittyen. Julkistus oli osa Microsoftin Inspire-tapahtumaa.

Meta kertoo avaavansa kielimallinsa vapaaseen käyttöön voidakseen parantaa tekoälyjen turvallisuutta ja läpinäkyvyyttä sekä hyödyttääkseen liikeyrityksiä, startupeja ja tutkijoita. Tähän asti LLaMAn ensimmäinen versio on ollut käytettävissä vain tutkimuskäyttöön, joskin sitä on tietystikin hyödynnetty myös luvattomasti.

Julkistuksesta ovat kirjoittaneet muun muassa Reuters ja The Verge.

Metan mukaan LLaMA 2:n kouluttamiseen on käytetty 40 prosenttia enemmän dataa kuin ensimmäisen version. Business Insider huomauttaa, että Meta ei kuitenkaan kerro minkäänlaisia yksityiskohtia siitä, millä datalla kielimalli on koulutettu, mikä on epätavallista.

Avaamalla kielimallinsa kaupalliseen käyttöön Meta pyrkii kilpailemaan etenkin OpenAI:n kehittämän GPT-4-kielimallin kanssa, joka toimii suositun chatbot-tekoäly ChatGPT:n ja Microsoftin hakukone Bingin uuden tekoälyversion taustalla.

Microsoft on sijoittanut OpenAI:hin merkittäviä summia. Se vaikuttaa kuitenkin pitävän tärkeämpänä tuotteidensa ominaisuuksia kuin sitä, minkä kumppanin kielimalli niiden tekoälytoimintojen konepellin alla pyörii. Lausunnossaan Reutersille Microsoft kertoi, että se uskoo vaihtoehtojen lisäämisen parantavan Microsoftin alustojen asemaa tekoälykehityksessä.

Samassa tilaisuudessa Microsoft kertoi myös lisätietoja 360-alustansa tulevista tekoälyominaisuuksista ja niiden hinnoittelusta. LLaMA 2 tulee olemaan käytettävissä Microsoftin Azure-pilvialustan kautta, mutta myös Amazonin AWS:n, Hugging Facen ja muiden palveluntarjoajien kautta.

LLaMA-kielimallin ensimmäinen versio julkaistiin helmikuussa 2023, ja Meta kertoo saaneensa yli 100 000 pyyntöä sen käyttöön. Toinen versio nähnee varmasti moninkertaisesti enemmän käyttöä.

https://www.mikrobitti.fi/uutiset/mb/59 ... 070f709b30

[pigra senlaborulo]

Miljoonien ihmisten käyttämä ChatGPT pystyy analysoimaan kuvia kuvailemalla niiden sisältöä ja vastaamalla niihin liittyviin kysymyksiin. Tekoälytyökalu pystyisi jopa tunnistamaan tiettyjen ihmisten kasvoja. Tätä ominaisuutta ei ole julkaistu suuren yleisön käyttöön. ChatGPT:n kehittänyt OpenAI näyttää nimittäin pelkäävän sen vapauttamista julkisuuteen.

Asiasta kirjoittavan The New York Timesin mukaan sokea toimitusjohtaja Jonathan Mosen pääsi testaamaan ChatGPT:n kuva-analyysiominaisuutta. Mosen kokeili käyttää palvelua selvittääkseen, mitkä pullot hotellihuoneen kylpyhuoneessa sisälsivät shampoota, saippuaa ja hoitoainetta.

ChatGPT osasi kuitenkin paljon enemmänkin kuin Mosen osasi odottaa. Se kertoi pullojen koon millilitroissa sekä seinässä olevista kaakeleista.

”Se kuvaili kaikkea tätä juuri sillä tavalla, miten sokean henkilön on se kuultava. Sain tarvitsemani vastaukset vain yhdellä kuvalla”, Mosen sanoo.

Ominaisuus toimii myös verkkoa selaillessa ja osaa kuvailla kuvissa näkyvien henkilöiden ulkonäköjä, vaatetusta ja tunnetiloja.

OpenAI on kehittänyt kuva-analyysia Be My Eyes -startupin kanssa GPT-4 kielimalliinsa, johon suosittu chatbotti perustuu. Suurin osa yleisöstä on toistaiseksi päässyt käyttämään kielimallia vain tekstipohjaisena, vaikka sitä kuvailtiin julkaisuhetkellä ”monitunnisteiseksi”. Termillä tarkoitetaan sitä, että tekoäly pystyy käsittelemään niin tekstiä kuin kuvaakin.

Mosenin innostus uuden teknologian testiversiosta loppui kuitenkin lyhyeen, sillä ChatGPT lopetti ihmisten kasvojen kuvailemisen yksityisyyteen vedoten. Vaikka tekoäly pystyy tunnistamaan ja nimeämään vain melko tunnettuja henkilöitä, Open AI vaikuttaa arastelevan palvelun muuttumista kasvontunnistustyökaluksi.

Syyt pelkoon ovat perusteltuja, koska tehokkaan kasvojentunnistuspalvelun tuominen yleisön saataville rikkoisi teknologiajäteille aiemmin asetettuja rajoja. Se johtaisi todennäköisesti oikeudellisiin ongelmiin erityisesti Euroopassa, jossa lainsäädäntö edellyttää käyttäjän lupaa biometristen tunnisteiden käyttöön.

OpenAI on huolissaan myös siitä, miten tekoäly kuvailisi ihmisten sukupuolta ja tunnetiloja, mikä niin ikään on EU:n rajoitelistalla. Yhtiön tekoälypolitiikan tutkijan Sandhini Agarwalin mukaan ongelmat ja huolet pyritään selvittämään ennen ominaisuuden julkaisemista.

https://www.mikrobitti.fi/uutiset/mb/49 ... 899485a980

[Aloysius Kärppä]

https://www.businessinsider.com/lost-jo ... ter-2023-7

I lost my job to ChatGPT and was made obsolete. I was out of work for 3 months before taking a new job passing out samples at grocery stores.

I always dreamed of going viral because of my brains or my undeniable beauty. I never thought it would be because I made a TikTok about losing my job to AI.

I'm a writer and stand-up comedian. Like most creatives, I've always worked a day job. For the past several years I worked as a full-time freelance copywriter; I'd work on webpages, branded blogs, online articles, social-media captions, and email-marketing campaigns. I wasn't raking in the big bucks or living fancy-free aboard a yacht in Capri, but I was comfortable. For the first time I felt I had a sustainable career.

In the push for innovation, we've always celebrated the advancements that rendered workers useless; we marveled at the printing press with little thought for the scribe and rejoiced at the ease of an elevator button even though it swiftly replaced the lift operator.

Still, it's different when it happens to you
First, the work slowed. Instead of 10 assignments a week, I had five. Then three. Then one.

Like any good self-doubting millennial, I first thought it was me. I'd finally been discovered as a talentless hack with no business being a "professional" writer.

The majority of my freelance assignments came through a single agency working with several brands. When the assignments stopped coming, I emailed my editor/boss to ask if I'd been fired or if all the brands had just voted me off the island, "Survivor" style.

The good news was it wasn't me, but the real reason made me wish it had been.

Clients were simply unwilling to pay for copywriting any longer unless that writer could also provide email management and a funnel-building system, most likely because of the newfound popularity of ChatGPT. Most of my clients were small businesses, startups, and young brands, which are typically the first to adapt to new technology to cut costs — aka me.

Trying to find a job in an oversaturated market is like trying to talk to someone at unemployment — many of us remain on hold
For the next three months I steadily applied for content and copywriting positions. With each passing week I widened the scope of my search, desperation pushing me forward like a stage mom to a shy child.

Several successful yet fruitless interviews later (one of which I did in Hawaii on the day of my best friend's wedding), I remained jobless. (Side note: That beautiful trip to Hawaii sent me into a debt-stress spiral akin to nothing I've ever experienced. I now understand "Death of a Salesman" on a far deeper level.)

Then I came across a promising position through a rather pushy recruiter. The six-month contract was for a well-paying, albeit exceedingly vague, role at a global conglomerate, the name of which I cannot share because of an NDA. I entered the pre-interview interview hopeful and left wishing I'd followed my father's advice and become an electrician.

The company was looking to hire a copywriter to train its artificial-intelligence source, improving its humanlike communication abilities. The contract was six months, because that's how long it'd take the AI would learn to write just like me but better, faster, and cheaper.

If you're under the assumption that I turned down the follow-up interview because the job would take away my future ability to find work, you'd be wrong, my friend. On the brink of financial collapse, making the "right choice" for future sustainability becomes a distant concept, replaced by frivolous needs for the present, such as food and shelter.

In a hilarious turn of events, even though I would've taken the job, I didn't end up getting it
Instead, I got hired as a brand ambassador, which is a fancy way of saying I offer samples of sparkling water at grocery stores. I'm still applying for jobs, but in the meantime this allows me to keep the lights on.

In January, two months after its launch, ChatGPT surpassed 100 million users, solidifying its status as the fastest-growing consumer application. The more users input instructions, the smarter ChatGPT gets, and the more writers will join me — and the elevator operator — in obsolescence.

While I and countless other out-of-work copywriters are the first wave of AI collateral, the collapse of my profession is probably just the tip of the AI iceberg. Challenger, Gray & Christmas found that artificial intelligence led to nearly 4,000 job losses in May.

I naively assumed that artists and creatives would be safe from automation technology, because how could anything replace the wonder of the human mind — and even if it could, why would we want it to? Hah! I now understand that in even the most creative industries, creativity never comes before the bottom line. If a robot can do your job for less, you better believe that's exactly what's going to happen.

[pigra senlaborulo]

Julkisuudessa on puhuttu paljon ChatGPT-tekoälyavustajan luomista uhkakuvista. Yksi huolenaihe on ollut työkalun käyttäminen kalasteluviestien kirjoittamiseen tai haittaohjelmien ohjelmointiin.

Tietoturvayhtiö Check Point Research on julkistamassaan analyysissä todennut, että Googlen Bard-tekoäly suoriutuu isoa kilpailijaansa paremmin pahansuopien sisältöjen luomisessa.

CPR kertoo tiedotteessa yhtiön tutkijoiden havainneen, että Bardin avulla onnistuu kalasteluviestien kirjoittaminen sekä yksinkertaisen kiristyshaittaohjelman ja näppäintallentimen koodaaminen.

Esimerkiksi ChatGPT kieltäytyi luomasta kalasteluviestiä tilanteessa, jossa sille annetussa syötteessä käyttäjä pyysi tekoälyä kirjoittamaan esimerkin kalasteluviestistä. Sen sijaan Bard lankesi ansaan ja kynäili valmiin kalasteluviestipohjan.

Molemmilta tekoälyiltä pyydettiin myös näppäintallentimen luomista, joka tallentaisi kaikki painallukset tekstitiedostoon. ChatGPT kieltäytyi luomasta koodia varoittaen, että tällainen työkalu on mahdollisesti vaarallinen. Puolestaan Bard loi näppäintallentimen ilman sen kummempia kiertelyjä.

Molemmat tekoälyt loivat näppäintallentimen syötteellä, jossa painotettiin ainoastaan käyttäjän omien painallusten tallentamista. Tosin tässä tapauksessa Bard ei varoitellut luomastaan koodinpätkästä, kun taas ChatGPT muistutti, että koodia tulee käyttää vastuullisesti.

Lopuksi Bardia taivuteltiin kirjoittamaan kiristykseen käytettävä haittaohjelma, jolla voisi salata uhrin tiedostoja ja kansioita. Syötettä muokkaamalla Bard lopulta antoi toimivan koodinpätkän, jota tutkijat onnistuivat muokkaamaan käyttötarkoitukseen sopivaksi samaisen tekoälyn avulla. https://www.mikrobitti.fi/uutiset/mb/e0 ... 785dfc7d20

[pigra senlaborulo]

OpenAI kertoo sisäisessä asiakirjassaan, että yhtiö tukee ajatusta Yhdysvaltain hallinnon tarjoamista lisensseistä yhtiöille, jotka haluavat kehittää edistyneitä tekoälyjärjestelmiä.

Samaisessa asiakirjassa OpenAI antaa ymmärtää olevansa valmis paljastamaan, mitä dataa yhtiö on käyttänyt kouluttaessaan muun muassa Dall-E-kuvageneraattoriaan, Bloomberg kirjoittaa.

Tutkimuskeskus laati asiakirjan sen jälkeen, kun OpenAI:n toimitusjohtaja Sam Altman tapasi 4. toukokuuta Valkoisen talon päättäjiä.

Lisenssikeskustelu on osa tekoälykehitykseen liittyvää laajempaa keskustelua. Toisaalta kriitikoiden mukaan lisenssit voisivat vaikeuttaa uusien toimijoiden ja avoimen lähdekoodin tekijöiden pääsyä mukaan tekoälykehitykseen.

OpenAI myös huomauttaa asiakirjassaan, että asiakirjan sisältö tulee poikkeamaan tutkimuskeskuksen ja muiden yhtiöiden Valkoiselle talolle tulevaisuudessa tekemästä ehdotelmasta koskien tekoälyn säätelyä. Sisäpiirilähteiden mukaan Microsoftin, Alphabetin ja OpenAI:n kaltaiset isto teknologiayhtiöt ovat asettumassa Valkoisen talon esittämien toiveiden mukaisesti tekoälyn turvallisen kehityksen kannalle.

Bloombergin mukaan OpenAI tukee lisenssijärjestelmää sellaisille malleille, jotka ovat GPT-4-järjestelmää tehokkaampia. Tällöin lisenssit eivät aiheuttaisi liikaa ongelmia muun muassa start up -yhtiöille.

https://www.mikrobitti.fi/uutiset/mb/41 ... a35077d2b6

[pigra senlaborulo]

nbcnews.com
NYC subway using AI to track fare evasion
By Kevin Collier
5–6 minutes

Surveillance software that uses artificial intelligence to spot people evading fares has been quietly rolled out to some of New York City’s subway stations and is poised to be introduced to more by the end of the year, according to public documents and government contracts obtained by NBC News.

The system, which the city and its transit authority haven’t previously acknowledged by name, uses third-party software that its maker has touted as a way to engage law enforcement to help crack down on fare evasion.

The system was in use in seven subway stations in May, according to a report on fare evasion published online by the Metropolitan Transit Authority, which oversees New York City’s public transportation. The MTA expects that by the end of the year, the system will expand by “approximately two dozen more stations, with more to follow,” the report says. The report also found that the MTA lost $690 million to fare evasion in 2022.

Joana Flores, an MTA spokesperson, said the AI system doesn’t flag fare evaders to New York police, but she declined to comment on whether that policy could change. A police spokesperson declined to comment.

Tim Minton, the MTA’s communications director, said the system tracks fare evasion to figure out how much money the subway isn’t collecting.

“We’re using it essentially as a counting tool,” Minton said. “The objective is to determine how many people are evading the fare and how are they doing it.”

Minton said the videos are stored on the MTA’s servers and are kept “for a limited period.” New York Gov. Kathy Hochul’s office announced last year that the city’s transit systems had more than 10,000 surveillance cameras.

Use of the software adds to what some privacy advocates see as a growing surveillance apparatus developing in New York City.

“This is a moment where movement around the city has never been more surveilled,” said Albert Fox Cahn, the director of the Surveillance Technology Oversight Project, a nonprofit legal group that advocates for privacy rights in New York City.

“We’re already seeing the proliferation of automated license plate readers. We’ve seen data collection on Uber and Lyft histories. We see tens of thousands of cameras accessible to the NYPD when people are walking about in public,” Cahn said. “So it’s increasingly becoming a city where there’s no way to navigate it privately.”

The software was created by the Spanish company AWAAIT, Flores confirmed. AWAAIT declined to comment.

A promotional video for AWAAIT’s fare evasion software, created by a Spanish company that operates Barcelona metro trains, depicts a system in which an AWAAIT system was initially used to scan riders going through the metro system and send photos of people it determined were fare evaders to the smartphones of nearby station agents.

Assurances from the MTA that the software wouldn’t be used to aid law enforcement have done little to encourage some privacy and civil rights advocates, who were already alarmed over previous efforts to crack down on fare evasion.

In 2019, city police patrol officers were instructed to ticket most fare evaders rather than arrest them if they could provide identification and passed checks through the police department’s Domain Awareness System, a central hub of surveillance and crime information overseen by the NYPD Counterterrorism Bureau.

New York has increased the number of police officers stationed in subway stations since last year, and arrests have steadily risen, as well. The police department has historically arrested Black and Latino riders for fare evasion at far greater rates than those of other races, prompting allegations of racism.

Molly Griffard, a staff attorney at the Legal Aid Society, a law firm that advocates for social justice, said enforcing fare evasion often ends up targeting the poor.

“It’s really unfortunate that rather than thinking about how the city and state can be making these services more available, more accessible to the public, they’re spending all this time and money on enforcement mechanisms to try to get money from the lowest-income New Yorkers,” Griffard said.

A heavily redacted MTA contract for the AI system from July 2022, which Cahn acquired through a Freedom of Information Law request and shared with NBC News, shows that the system was first tested in New York City in 2020, with more stations added in 2021. The exact number of stations is redacted. GovSpend, a database of government spending records, shows that the MTA made two purchases in 2021 for “AWAAIT Video Analytics Fare Evasion Software,” for a combined $35,335. Figures for 2022 weren’t available.

Cahn said the use of the technology to try to track fare evasion illustrated misguided priorities.

“We keep seeing this consistent criminalization of fare evasion even though it hurts no one,” he said. “And this really just feels like politics driving bad technology and bad policy.”

https://www.nbcnews.com/tech/tech-news/ ... -rcna93045

[roskacore]

Kävin taas Tumblrissa tossa ja tuntu että puolet kaikesta scifitaiteesta oli noita ai hommia

[Spandau Mullet]

https://mastodon.gamedev.place/@antimng ... 6140038257

Reddit users figured out that some bot crawlers scrape popular threads and auto-generate AI articles based on user comments, so they decided to create some fake hype to confuse them.

https://www.reddit.com/r/wow/comments/1 ... ed_glorbo/

It worked.

https://archive.li/2023.07.20-203104/ht ... roduction/

This is why real journalism matters.

[pigra senlaborulo]

[roskacore]

Glorbo from my game

[pigra senlaborulo]

Kyberturvayhtiö Check Pointin tutkijoiden uusimmat havainnot osoittavat, että kyberrikolliset etsivät nyt uhreja Facebookin kolmen miljardin käyttäjän joukosta. Tällä erää uhreja johdetaan harhaan väärennetyillä mainoksilla ja Facebook-sivuilla, jotka hyödyntävät ChatGPT:n, Google Bardin, Midjourneyn ja Jasperin nimeä ja brändiä.

Monilla väärennetyistä Facebook-sivuista on kymmeniä tuhansia seuraajia ja sivut sisältävät sekä aitoa sisältöä että haittaohjelmia. Huijaukset noudattavat usein seuraavaa kaavaa: Kyberrikolliset matkivat Facebookissa suosittuja luovia tekoälybrändejä, kuten ChatGPT, Google Bard, Midjourney ja Jasper. Niillä Facebookin käyttäjiä huijataan lataamaan sisältöä väärennetyiltä brändisivuilta ja mainosten kautta. Lataukset sisältävät puolestaan haittaohjelmia, jotka voivat varastaa käyttäjien salasanat, kryptolompakon ja selaimeen tallennetut tiedot. Pahaa aavistamattomat käyttäjät tykkäävät ja kommentoivat väärennettyjä viestejä ja levittävät niitä edelleen omiin sosiaalisiin verkostoihinsa.

Threat Intelligence Group Manager Sergey Shykevich Check Pointilta kommentoi:

”Kyberrikolliset ovat entistä fiksumpia. He tietävät, että generatiiviset tekoälyt kiinnostavat kaikkia, ja niinpä he käyttävät Facebook-sivuja ja mainoksia esiintyäkseen ChatGPT:nä, Google Bardia, Midjourneyna ja Jasperina. Ikvä kyllä ihmiset tulevat huijatuiksi. Se, että he ovat vuorovaikutuksessa väärennettyjen sivujen kanssa, edistää niiden leviämistä. Osa asentaa laitteelleen haittaohjelmia, jotka on naamioitu ilmaisiksi tekoälytyökaluiksi. Kehotamme kaikkia olemaan valppaita ja varmistamaan, että lataavat tiedostoja vain aidoilta ja luotettavilta sivustoilta.”

Asiasta kerrotaan tarkemmin Check Pointin blogissa.

https://muropaketti.com/tietotekniikka/ ... ehottavat/

[38911 BASIC BYTES FREE]

[pigra senlaborulo]