Artificial intelligence is rapidly changing many aspects of how we work and live. (How many stories did you read last week about self-driving cars and job-stealing robots? Perhaps your holiday shopping involved some AI algorithms, as well.) But despite the constant flow of news, many misconceptions about AI remain.
AI doesn’t think in our sense of the word at all, Scriffignano explains. “In many ways, it’s not really intelligence. It’s regressive.”
IT leaders should make deliberate choices about what AI can and can’t do on its own. “You have to pay attention to giving AI autonomy intentionally and not by accident,”
5,954 Comments
Tomi Engdahl says:
Benedict Evans:
A look at the ethical and legal issues around generative AI, which makes things that were previously only possible on a small scale practical at a massive scale — We’ve been talking about intellectual property in one way or another for at least the last five hundred years
Generative AI and intellectual property
https://www.ben-evans.com/benedictevans/2023/8/27/generative-ai-ad-intellectual-property
If you put all the world’s knowledge into an AI model and use it to make something new, who owns that and who gets paid? This is a completely new problem that we’ve been arguing about for 500 years.
Tomi Engdahl says:
Gillian Tett / Financial Times:
How LLMs are impacting computational humor, an AI subfield that uses computers to create jokes, as some comedians begin using AI chatbots for improv and roasts
Can AI crack comedy?
https://www.ft.com/content/818f2cab-57ff-42c3-917b-4a83f1d87802
Tomi Engdahl says:
Mia Sato / The Verge:
A look at Smart Answers, an AI chatbot trained only on Macworld, PCWorld, Tech Advisor, and TechHive content, and offered to the outlets’ readers on August 1
https://www.theverge.com/2023/8/25/23844868/ai-chatbot-macworld-pcworld-journalism-smart-answers
Tomi Engdahl says:
Amir Efrati / The Information:
Source: OpenAI is on pace to generate more than $1B in revenue over the next 12 months by selling software and computing capacity, up from just $28M in 2022 — OpenAI is currently on pace to generate more than $1 billion in revenue over the next 12 months from the sale of artificial intelligence software …
OpenAI Passes $1 Billion Revenue Pace as Big Companies Boost AI Spending
By Amir Efrati
and Aaron Holmes
https://www.theinformation.com/articles/openai-passes-1-billion-revenue-pace-as-big-companies-boost-ai-spending
| Aug. 29, 2023 3:58 PM PDT
Photo: OpenAI CEO Sam Altman. Photo via Getty
OpenAI is currently on pace to generate more than $1 billion in revenue over the next 12 months from the sale of artificial intelligence software and the computing capacity that powers it. That’s far ahead of revenue projections the company previously shared with its shareholders, according to a person with direct knowledge of the situation.
The billion-dollar revenue figure implies that the Microsoft-backed company, which was valued on paper at $27 billion when investors bought stock from existing shareholders earlier this year, is generating more than $80 million in revenue per month. OpenAI generated just $28 million in revenue last year before it started charging for its groundbreaking chatbot, ChatGPT. The rapid growth in revenue suggests app developers and companies—including secretive ones like Jane Street, a Wall Street firm—are increasingly finding ways to use OpenAI’s conversational text technology to make money or save on costs. Microsoft, Google and countless other businesses trying to make money from the same technology are closely watching
Tomi Engdahl says:
AI Assistant Translates Your Every Request For The Command Line
https://hackaday.com/2023/08/30/ai-assistant-translates-your-every-request-for-the-command-line/
If you don’t live on the command line, it can be easy to forget the exact syntax of commands. It often leaves you running to the “/?” or “–help” switches, or else a quick Google search to find the proper incantations. Shell-AI is a machine-learning assistant that could change all that by helping you find the proper command for the job, right on the command line!
Shell-AI accepts natural-language inputs — simply type in “shai” followed by what you’re trying to do. It will then take in your request, run it through an OpenAI language model like GPT-3.5-Turbo, and then present you with three (or more) potential commands. You can then select which command to use and get on with your day.
As demonstrated, it’s more than capable of following commands like “download a random image” or “show only image files ls.” And, hilariously, it responds to the request “do something crazy” with just one suggestion: “rm -rf”. That seems rather fitting.
Shell-AI: Your Intelligent Command-Line Companion
https://github.com/ricklamers/shell-ai
Shell-AI (shai) is a CLI utility that brings the power of natural language understanding to your command line. Simply input what you want to do in natural language, and shai will suggest single-line commands that achieve your intent. Under the hood, Shell-AI leverages the LangChain for LLM use and builds on the excellent InquirerPy for the interactive CLI.
Tomi Engdahl says:
AI Lends a Helping Hand with Analog and Custom IC Design
Aug. 28, 2023
4
Siemens’ new Solido Design Environment highlights the role that artificial intelligence is playing in the world of EDA.
https://www.electronicdesign.com/technologies/embedded/article/21272567/electronic-design-ai-lends-a-helping-hand-with-analog-and-custom-ic-design?utm_source=EG+ED+Analog+%26+Power+Source&utm_medium=email&utm_campaign=CPS230824101&o_eid=7211D2691390C9R&rdx.identpull=omeda|7211D2691390C9R&oly_enc_id=7211D2691390C9R
Skilled engineers remain the driving force for innovation in chips. But it’s no secret that electronic design automation (EDA) companies are folding AI into more of their offerings to speed up design and verification.
But as industry insiders tell it, these AI-powered EDA tools don’t have enough intelligence to actually replace human designers anytime soon. Instead, the biggest difference these tools are having is on the productivity front, with AI reducing the number of hours that engineers spend on the more tedious phases of the design and verification process, said Amit Gupta, VP and GM of Siemens Digital Industries Software’s custom IC verification division.
Tomi Engdahl says:
https://hackaday.com/2023/08/27/text-compression-gets-weirdly-efficient-with-llms/
Tomi Engdahl says:
https://www.uusiteknologia.fi/2023/08/29/suomalainen-supertietokone-jauhaa-avointa-kielimallia-eurooppaan/
Tomi Engdahl says:
Tekoälyprosessoreja myydään tänä vuonna jo 53 miljardilla dollarilla
https://etn.fi/index.php/13-news/15249-tekoaelyprosessoreja-myydaeaen-taenae-vuonna-jo-53-miljardilla-dollarilla
Tekoälyprosessoreja myydään tänä vuonna jo 53 miljardilla dollarilla
Julkaistu: 28.08.2023
Devices Embedded Business
Tekoälylaskentaa suorittavien prosessorien myynti kasvaa tänä vuonna 53,4 miljardiin dollariin tänä vuonna, arvioi Gartner. Kasvua tulee viime vuodesta 20,9 prosenttia ja se perustuu generatiivisen tekoälyn nopeaan laajenemiseen datakeskuksissa, reunalaitteissa ja päätelaitelaitteissa.
Gartnerin ennusteen mukaan tekoälypuolijohteiden myynti kasvaa ensi vuonna 67,1 miljardiin dollariin eli kasvua kiihtyy 25,6 prosenttiin. Käytännössä AI-piirien osuus koko puolijohdemarkkinasta kasvaa jo erittäin merkittäväksi ja esimerkiksi PC-prosessoreja suuremmaksi.
Vuoteen 2027 mennessä AI-sirujen liikevaihdon odotetaan saavuttavan 119,4 miljardia dollaria.
Gartner ennustaa myös, että tekoälyprosessorit alkavat löytää tietään myös kulutuselektroniikan laitteisiin.
Tomi Engdahl says:
Google Workspace Introduces New AI-Powered Security Controls
https://www.securityweek.com/google-workspace-introduces-new-ai-powered-security-controls/
Google has announced new AI-powered zero trust, digital sovereignty, and threat defense controls for Workspace customers.
Google on Thursday introduced new AI-powered security controls for its Workspace customers, targeting zero trust, digital sovereignty, and threat defense.
The new AI-powered zero trust capabilities, Google says, are meant to provide organizations with more granular control over how data is accessed and used.
To ensure data protection and prevent inappropriate sharing of data in Google Drive, Google AI can now be used to automatically and continuously classify and label new and existing files, and then apply necessary controls based on the organization’s security policies.
Context-aware DLP controls will allow administrators to set specific criteria to be met before a user can share sensitive content in Drive – the capability will become available in preview later this year.
Gmail will receive enhanced DLP controls too – also in preview later this year – to improve control over the sharing of sensitive information, both inside and outside the organization.
The internet giant also announced new digital sovereignty controls to help prevent unauthorized access to sensitive data, storing encryption keys, selecting where data is processed, and limiting Google support access.
Tomi Engdahl says:
A Bard’s Tale – how fake AI bots try to install malware
The AI race is on! It’s easy to lose track of the latest developments and possibilities, and yet everyone wants to see firsthand what the hype is about. Heydays for cybercriminals!
https://www.welivesecurity.com/en/scams/a-bards-tale-how-fake-ai-bots-try-to-install-malware/
Tomi Engdahl says:
ICYMI: OpenAI finally responded to a pair of class-action lawsuits from book authors—including Sarah Silverman—disputing claims that ChatGPT was illegally trained on pirated copies of their books: “The use of copyrighted materials by innovators in transformative ways does not violate copyright.” https://trib.al/VGqsJLH
Tomi Engdahl says:
Virheettömästä Milla-Sofiasta tuli Tiktok-tähti – tilin takaa paljastuu 44-vuotias mies Vantaalta
Tutkijat ovat erimielisiä siitä, miten luovalla tekoälyllä tehtyihin kuviin pitäisi suhtautua. Niiden käyttö lisääntyy kuitenkin koko ajan.
https://yle.fi/a/74-20046003?origin=rss&fbclid=IwAR3LwtiyaZCcXSza1DXL7XAeVO9AfNPcaIyPnPHE45z9kaEwBP4-5Fnh8zU
Tomi Engdahl says:
Prompt injection could be the SQL injection of the future, warns NCSC https://www.malwarebytes.com/blog/news/2023/08/prompt-injection-could-be-the-sql-injection-of-the-future-warns-ncsc
The UK’s National Cyber Security Centre (NCSC) has issued a warning about the risks of integrating large language models (LLMs) like OpenAI’s ChatGPT into other services. One of the major risks is the possibility of prompt injection attacks.
The NCSC points out several dangers associated with integrating a technology that is very much in early stages of development into other services and platforms. Not only could we be investing in a LLM that no longer exists in a few years (anyone remember Betamax?), we could also get more than we bargained for and need to change anyway.
Tomi Engdahl says:
Kate Knibbs / Wired:
Copyright activists are working to wipe Books3 from the internet, which may only benefit the big companies that have already been using the AI training dataset
The Battle Over Books3 Could Change AI Forever
https://www.wired.com/story/battle-over-books3/
Copyright activists are on a mission to wipe a popular generative AI training set from the internet. Success could alter the industry—and who controls it.
Tomi Engdahl says:
Christopher Mims / Wall Street Journal:
How startups offering AI-powered cameras, tree-mounted sensors, water-dumping drones, satellite data, and more are helping prevent, detect, and fight wildfires
How Better Tech Could Save Lives in a World of Bigger, Faster, More Devastating Fires
We can already detect fires from space, soon after they start. Here’s why we don’t yet have a nationwide system for alerting us when they do—but could someday.
https://www.wsj.com/tech/ai/how-better-tech-could-save-lives-in-a-world-of-bigger-faster-more-devastating-fires-665c7fb5?mod=followamazon
Tomi Engdahl says:
Noah Smith / Noahpinion:
How LLMs like GPT-4 can have an equalizing effect in some cases, boosting less-skilled workers and potentially leveling the human-capital playing field
Is it time for the Revenge of the Normies?
An optimistic take on technology and inequality.
https://www.noahpinion.blog/p/is-it-time-for-the-revenge-of-the
Tomi Engdahl says:
A.I.’s un-learning problem: Researchers say it’s virtually impossible to make an A.I. model ‘forget’ the things it learns from private user data
https://fortune.com/2023/08/30/researchers-impossible-remove-private-user-data-delete-trained-ai-models/
Tomi Engdahl says:
While AI has previously bested humans in games like chess, Go, and even StarCraft, this may be the first time an AI system has outperformed human pilots in a physical sport.
High-speed AI drone beats world-champion racers for the first time
University creates the first autonomous system capable of beating humans at drone racing.
https://arstechnica.com/information-technology/2023/08/high-speed-ai-drone-beats-world-champion-racers-for-the-first-time/?utm_source=facebook&utm_brand=ars&utm_medium=social&utm_social-type=owned&fbclid=IwAR2hRCq9jDsx8LMklXy08CVvnZmz2sxt9cvYV50XXeRahrduMUkXzDSWBNs
Tomi Engdahl says:
https://hackaday.com/2023/09/03/heres-why-gpus-are-deep-learnings-best-friend/
Tomi Engdahl says:
https://hackaday.com/2023/09/01/re-creating-pink-floyd-in-the-name-of-speech/
Tomi Engdahl says:
High Quality 3D Scene Generation From 2D Source, In Realtime
https://hackaday.com/2023/09/02/high-quality-3d-scene-generation-from-2d-source-in-realtime/
Here’s some fascinating work presented at SIGGRAPH 2023 of a method for radiance field rendering using a novel technique called Gaussian Splatting. What’s that mean? It means synthesizing a 3D scene from 2D images, in high quality and in real time, as the short animation shown above shows.
Neural Radiance Fields (NeRFs) are a method of leveraging machine learning to, in a way, do what photogrammetry does: synthesize complex scenes and views based on input images. But NeRFs work in a fraction of the time, and require only a fraction of the source material. There are different ways to go about this and unsurprisingly, there tends to be a clear speed vs. quality tradeoff. But as the video accompanying this new work seems to show, clever techniques mean the best of both worlds.
3D Gaussian Splatting
for Real-Time Radiance Field Rendering
https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/
Tomi Engdahl says:
Tekoäly valitsee yhä useammin komponentit suunnitteluun
https://etn.fi/index.php/13-news/15289-tekoaely-valitsee-yhae-useammin-komponentit-suunnitteluun
Suunnittelijoiden komponenttikauppias Farnellin tuore tutkimus osoittaa, että insinöörit luottavat nyt tekoälyyn valitessaan komponentteja uusiin suunnitteluihinsa. Peräti 86 prosenttia suunnittelijoista luottaa tekoälyyn ainakin jossain määrin komponenttien valinnassa.
Lisäksi lähes joka neljäs vastaaja (23 prosenttia) sanoi, että he mielellään luottaisivat “täysin” tekoälyyn komponenttien valinnassa. Kyselyn tulokset kuitenkin osoittivat, että vaikka insinöörit uskovat, että tekoälyllä on kasvava rooli komponenttien valinnassa tulevaisuudessa, on olemassa jatkuva huoli tekoälyjärjestelmien tahallisesta tai tahattomasta harhasta.
Jotkut suunnittelijat sanoivat luottavansa tekoälyyn, kunhan niillä on “rajoitettu” rooli komponenttien valinnassa. Vaikka useimmat vastaajat pitivät täydentävää tekoälyä tervetulleina, he myös katsoivat, että ihmisiä tarvitaan aina osana valintaprosessia erityisesti turvallisuuskriittisten järjestelmien ja innovatiivisten suunnittelujen osalta.
Tomi Engdahl says:
Wayne Ma / The Information:
Source: Apple’s budget for training AI is now millions of dollars per day; its conversational AI team Foundational Models has ~16 people led by Ruoming Pang — Apple has been expanding its computing budget for building artificial intelligence to millions of dollars a day.
Apple Boosts Spending to Develop Conversational AI
https://www.theinformation.com/articles/apple-boosts-spending-to-develop-conversational-ai
Tomi Engdahl says:
Joe Coscarelli / New York Times:
Ghostwriter, the anonymous creator who used AI to mimic Drake and The Weeknd, has met with record labels, Grammy organizers, and more, and releases a new song — The anonymous artist, who stirred conversation with the A.I. track “Heart on My Sleeve,” has been quietly consulting with executives, while also gunning for a Grammy.
https://www.nytimes.com/2023/09/05/arts/music/ghostwriter-whiplash-travis-scott-21-savage.html
Tomi Engdahl says:
Andrew Deck / Rest of World:
A test of ChatGPT’s ability to respond in underrepresented languages like Bengali and Kurdish found translation errors, fabricated words, and illogical answers — Outside of English, ChatGPT makes up words, fails logic tests, and can’t do basic information retrieval.
We tested ChatGPT in Bengali, Kurdish, and Tamil. It failed.
https://restofworld.org/2023/chatgpt-problems-global-language-testing/
Outside of English, ChatGPT makes up words, fails logic tests, and can’t do basic information retrieval.
Tomi Engdahl says:
Rachel Metz / Bloomberg:
OpenAI announces its first developer conference, OpenAI DevDay, on November 6 in San Francisco, and plans to livestream a keynote presentation — OpenAI will host its first developer conference this fall, as the ChatGPT-maker looks to bolster interest in its products while facing growing competition …
OpenAI to Host First Developer Conference in San Francisco
https://www.bloomberg.com/news/articles/2023-09-06/openai-to-host-first-developer-conference-in-san-francisco#xj4y7vzkg
Tomi Engdahl says:
Davey Alba / Bloomberg:
Google plans a November policy update requiring election advertisers to prominently disclose when their ads contain generative AI-based images, video, or audio — Audio, images and video on political advertising will need to be labeled … The policy update, which applies starting mid-November …
Google to Require ‘Prominent’ Disclosures for AI-Generated Election Ads
https://www.bloomberg.com/news/articles/2023-09-06/google-to-require-prominent-disclosures-for-ai-generated-election-ads#xj4y7vzkg
Audio, images and video on political advertising will need to be labeled
Alphabet Inc.’s Google will soon require that all election advertisers disclose when their messages have been altered or created by artificial intelligence tools.
The policy update, which applies starting mid-November, requires election advertisers across Google’s platforms to alert viewers when their ads contain images, video or audio from generative AI — software that can create or edit content given a simple prompt. Advertisers must include prominent language like, “This audio was computer generated,” or “This image does not depict real events” on altered election ads across Google’s platforms, the company said
Tomi Engdahl says:
Mariella Moon / Engadget:
Slack announces Slack AI, which can generate channel highlights, thread summaries, and search answers based on messages, with a pilot starting “this winter”
‘Slack AI’ will summarize your work chat starting this winter
Salesforce will pilot the generative AI capabilities it developed for Slack later this year.
https://www.engadget.com/slack-ai-will-summarize-your-work-chat-starting-this-winter-130046724.html
Tomi Engdahl says:
CNBC:
Tencent launches its Hunyuan AI model for business use and is integrating the model with its existing video conferencing and social media products
Tencent releases AI model for businesses as competition in China heats up
https://www.cnbc.com/2023/09/07/tencent-releases-ai-model-hunyuan-for-businesses-amid-china-competition.html
Tomi Engdahl says:
China, North Korea pursue new targets while honing cyber capabilities https://blogs.microsoft.com/on-the-issues/2023/09/07/digital-threats-cyberattacks-east-asia-china-north-korea/
In the past year, China has honed a new capability to automatically generate images it can use for influence operations meant to mimic U.S. voters across the political spectrum and create controversy along racial, economic, and ideological lines.
This new capability is powered by artificial intelligence that attempts to create high-quality content that could go viral across social networks in the U.S. and other democracies. These images are most likely created by something called diffusion-powered image generators that use AI to not only create compelling images but also learn to improve them over time.
Today, the Microsoft Threat Analysis Center (MTAC) is issuing Sophistication, scope, and scale: Digital threats from East Asia increase in breadth and effectiveness, as part of an ongoing series of reports on the threat posed by influence operations and cyber activity, identifying specific sectors and regions at heightened risk.
Tomi Engdahl says:
AI is Changing Open Access Science
ChatGPT took the world by storm, and its technology has only improved since. The potential that AI offers is shaking up entire industries, but how is it changing Open Access science?
https://mdpiblog.wordpress.sciforum.net/2023/09/05/ai-open-access/
Tomi Engdahl says:
FORMER GOOGLE EXEC WARNS AI COULD CREATE A DEADLY PLAGUE
https://futurism.com/the-byte/former-google-exec-ai-plague
Tomi Engdahl says:
Uutinen
Näin ChatGPT syntyi – kukaan ei täysin ymmärrä, miten kielimallit toimivat
Panu Räty7.9.202306:11TEKOÄLYTULEVAISUUDEN TEKNIIKATDIGITALISAATIO
Keskustelevat tekoälyjärjestelmät kuten ChatGPT lupaavat mullistaa niin työmarkkinat, yhteiskunnan rakenteet kuin koko maailmamme. Mutta miten nämä järjestelmät oikeastaan toimivat?
https://www.tivi.fi/uutiset/nain-chatgpt-syntyi-kukaan-ei-taysin-ymmarra-miten-kielimallit-toimivat/5cbb56ba-5557-4b70-8217-396a566a8cec
Tomi Engdahl says:
AMD’s CEO Lisa Su Reiterates “AI Priority”, Forecasts $150 Billion Revenue by 2027
https://wccftech.com/amd-ceo-lisa-su-reiterates-ai-priority-forecasts-150-billion-revenue-by-2027/
Tomi Engdahl says:
“We desperately need guardrails on this landslide of misattribution and misinformation.”
Author discovers AI-generated counterfeit books written in her name on Amazon
Amazon resisted a removal request, citing lack of “trademark registration numbers.”
https://arstechnica.com/information-technology/2023/08/author-discovers-ai-generated-counterfeit-books-written-in-her-name-on-amazon/?utm_social-type=owned&utm_brand=ars&utm_medium=social&utm_source=facebook&fbclid=IwAR3rsXIjHA7_1LNSHRXEtyBoe09TwlmlZKayYTg8w08lpncApdZXSuQmkGg
Upon searching Amazon and Goodreads, author Jane Friedman recently discovered a half-dozen listings of fraudulent books using her name, likely filled with either junk or AI-generated content. Both Amazon and Goodreads resisted removing the faux titles until the author’s complaints went viral on social media.
In a blog post titled “I Would Rather See My Books Get Pirated Than This (Or: Why Goodreads and Amazon Are Becoming Dumpster Fires),” published on Monday, Friedman detailed her struggle with the counterfeit books.
“Whoever’s doing this is obviously preying on writers who trust my name and think I’ve actually written these books,” she wrote. “I have not. Most likely they’ve been generated by AI.”
It’s a rising problem in a world where scammers game Amazon’s algorithm to make a quick buck on fraudulent sales. In February, Reuters did a profile on authors using ChatGPT to write e-books, selling them through Amazon. In June, Vice reported on an influx of dozens of AI-generated books full of nonsense that took over Kindle bestseller lists.
I Would Rather See My Books Get Pirated Than This (Or: Why Goodreads and Amazon Are Becoming Dumpster Fires)
https://janefriedman.com/i-would-rather-see-my-books-pirated/
Tomi Engdahl says:
Can’t lose what you never had: Claims about digital ownership and creation in the age of generative AI
https://www.ftc.gov/business-guidance/blog/2023/08/cant-lose-what-you-never-had-claims-about-digital-ownership-creation-age-generative-ai
Generative AI tools that produce output based on copyrighted or otherwise protected material may, nonetheless, raise issues of consumer deception or unfairness. That’s especially true if companies offering the tools don’t come clean about the extent to which outputs may reflect the use of such material. This information could be relevant to people’s decisions to use one tool or another. It’s not unusual for the FTC to sue when sellers deceive consumers about how products were made, such as with cases involving environmental claims. The information could also be relevant to business decisions to use such a tool for commercial purposes, given that the businesses could be liable if their use of the output infringes protected works.
Companies should keep in mind the following:
When offering digital products, you must ensure that customers understand the material terms and conditions, including whether they’re buying an item or just getting a license to use it. Unilaterally changing those terms or undermining reasonable ownership expectations can get you in trouble, too.
Selling digital items created via AI tools is obviously not okay if you’re trying to fool people into thinking that the items are the work of particular human creators.
When offering a platform for creators to develop and display their work, be clear and upfront about their rights to access and take this work with them, as well as how the work will be used and presented. Again, don’t change the terms later.
When offering a generative AI product, you may need to tell customers whether and the extent to which the training data includes copyrighted or otherwise protected material.
In the 1960s, four American musicians toured South America, pretending to be the Beatles, a scheme that worked until people saw their faces and heard them play. That kind of scam wouldn’t likely get off the ground today, but through a mixture of deepfakes, voice synthesis, and text generation, one could now create some fake, “long lost” Beatles music or footage and put it out in the world. At least Sir Paul McCartney, who has been using AI himself for artistic ends, would have the resources to deal with it. But many other artists would not be so fortunate if their work gets digitally faked or misused.
Tomi Engdahl says:
Warner Music signs a record deal with first AI pop singer
The digital singer, Noonoouri, released her debut single ‘Dominoes’ on September 1.
https://interestingengineering.com/science/warner-music-signs-a-record-deal-with-first-ai-pop-singer?utm_source=Facebook&utm_medium=content&utm_campaign=organic&utm_content=Sep06&fbclid=IwAR3bcqcWFM7sDlV96r7Tp6V_Nl_EU13ZnICmyByB_ake-h2cLbxgd3F1EYQ
Instagram influencer Noonoouri never ages, never gets tired and never sleeps. She also doesn’t eat anything and can grow a pack of abdominal muscles on a whim. She also only exists as a digital entity on the internet.
And she’s just released her debut single, ‘ Dominoes’, after signing a record deal with Warner Music Central Europe. Noonoouri is a first-of-its-kind digital influencer to secure a major record deal.
One X user wrote, “Looks and sounds like plastic.”
According to the Independent, the creators used artificial intelligence (AI) technology to create Noonoouri’s singing voice. The voice was based on a real singer but was altered to give Noonoouri her unique voice. The lyricists, singers, and musicians who have worked on the song will receive royalties and publishing splits just like musicians do in other non-AI-generated songs.
The song also features German DJ and producer Alle Farben and is available to listen here.
https://m.youtube.com/watch?time_continue=127&v=3Yi_KxMxvVg&embeds_referring_euri=https%3A%2F%2Finterestingengineering.com%2F&source_ve_path=Mjg2NjY&feature=emb_logo
Tomi Engdahl says:
Associated Press:
Research: ChatGPT uses an estimated 500ml of water for every five to 50 prompts; Microsoft disclosed the company’s water use spiked 34% YoY in 2022, Google 20% — The cost of building an artificial intelligence product like ChatGPT can be hard to measure. But one thing Microsoft-backed OpenAI needed …
Artificial intelligence technology behind ChatGPT was built in Iowa — with a lot of water
https://apnews.com/article/chatgpt-gpt4-iowa-ai-water-consumption-microsoft-f551fde98083d17a7e8d904f8be822c4
The cost of building an artificial intelligence product like ChatGPT can be hard to measure.
But one thing Microsoft-backed OpenAI needed for its technology was plenty of water, pulled from the watershed of the Raccoon and Des Moines rivers in central Iowa to cool a powerful supercomputer as it helped teach its AI systems how to mimic human writing.
As they race to capitalize on a craze for generative AI, leading tech developers including Microsoft, OpenAI and Google have acknowledged that growing demand for their AI tools carries hefty costs, from expensive semiconductors to an increase in water consumption.
Building a large language model requires analyzing patterns across a huge trove of human-written text. All of that computing takes a lot of electricity and generates a lot of heat. To keep it cool on hot days, data centers need to pump in water — often to a cooling tower outside its warehouse-sized buildings.
In its latest environmental report, Microsoft disclosed that its global water consumption spiked 34% from 2021 to 2022 (to nearly 1.7 billion gallons, or more than 2,500 Olympic-sized swimming pools), a sharp increase compared to previous years that outside researchers tie to its AI research.
“It’s fair to say the majority of the growth is due to AI,” including “its heavy investment in generative AI and partnership with OpenAI,” said Shaolei Ren, a researcher at the University of California, Riverside who has been trying to calculate the environmental impact of generative AI products such as ChatGPT.
In a paper due to be published later this year, Ren’s team estimates ChatGPT gulps up 500 milliliters of water (close to what’s in a 16-ounce water bottle) every time you ask it a series of between 5 to 50 prompts or questions. The range varies depending on where its servers are located and the season.
“Most people are not aware of the resource usage underlying ChatGPT,” Ren said. “If you’re not aware of the resource usage, then there’s no way that we can help conserve the resources.”
Google reported a 20% growth in water use in the same period, which Ren also largely attributes to its AI work. Google’s spike wasn’t uniform
In response to questions from The Associated Press, Microsoft said in a statement this week that it is investing in research to measure AI’s energy and carbon footprint “while working on ways to make large systems more efficient, in both training and application.”
“We will continue to monitor our emissions, accelerate progress while increasing our use of clean energy to power data centers, purchasing renewable energy, and other efforts to meet our sustainability goals of being carbon negative, water positive and zero waste by 2030,” the company’s statement said.
“We recognize training large models can be energy and water-intensive” and work to improve efficiencies, it said.
“So if you are developing AI models within Microsoft, then you should schedule your training in Iowa instead of in Arizona,” Ren said. “In terms of training, there’s no difference. In terms of water consumption or energy consumption, there’s a big difference.”
For much of the year, Iowa’s weather is cool enough for Microsoft to use outside air to keep the supercomputer running properly and vent heat out of the building. Only when the temperature exceeds 29.3 degrees Celsius (about 85 degrees Fahrenheit) does it withdraw water, the company has said in a public disclosure.
That can still be a lot of water, especially in the summer. In July 2022, the month before OpenAI says it completed its training of GPT-4, Microsoft pumped in about 11.5 million gallons of water to its cluster of Iowa data centers, according to the West Des Moines Water Works. That amounted to about 6% of all the water used in the district, which also supplies drinking water to the city’s residents.
In 2022, a document from the West Des Moines Water Works said it and the city government “will only consider future data center projects” from Microsoft if those projects can “demonstrate and implement technology to significantly reduce peak water usage from the current levels” to preserve the water supply for residential and other commercial needs.
Tomi Engdahl says:
Wall Street Journal:
Sources: Meta plans to begin training an LLM in Q1 2024 on its own infrastructure that it hopes will be roughly as capable as OpenAI’s GPT-4 — Parent of Facebook and Instagram wants artificial-intelligence system to be as capable as OpenAI’s most advanced model — Platforms is setting its sights on OpenAI.
Meta Is Developing a New, More Powerful AI System as Technology Race Escalates
https://www.wsj.com/tech/ai/meta-is-developing-a-new-more-powerful-ai-system-as-technology-race-escalates-decf9451?mod=followamazon
Parent of Facebook and Instagram wants artificial-intelligence system to be as capable as OpenAI’s most advanced model
The parent of Facebook and Instagram is working on a new artificial-intelligence system intended to be as powerful as the most advanced model offered by OpenAI, the Microsoft
-backed startup that created ChatGPT, according to people familiar with the matter. Meta aims for its new AI model, which it hopes to be ready next year, to be several times more powerful than the one it released just two months ago, dubbed Llama 2.
The planned system, details of which could still change, would help other companies to build services that produce sophisticated text, analysis and other output. It is the work of a group formed early this year by Meta Chief Executive Mark Zuckerberg to accelerate development of so-called generative AI tools that can produce humanlike expressions. Meta expects to start training the new AI system, known as a large language model, in early 2024, some of the people said.
Plans for the new model, which haven’t previously been reported, are part of Zuckerberg’s effort to assert Meta as a major force in the AI world after it fell behind rivals. Competition in the area has sharply intensified this year, spawning divergent views on everything from which business models are best to how the technology should be regulated.
Tomi Engdahl says:
Kashmir Hill / New York Times:
How Meta and Google held back their tech to recognize unknown people’s faces due to privacy worries, opening the door for startups like Clearview AI and PimEyes — Engineers at the tech giants built tools years ago that could put a name to any face but, for once, Silicon Valley did not want to move fast and break things.
https://www.nytimes.com/2023/09/09/technology/google-facebook-facial-recognition.html
Tomi Engdahl says:
Jay Peters / The Verge:
Roblox unveils Roblox Assistant, a conversational AI assistant that lets creators type in prompts to generate virtual environments, get coding help, and more
Roblox’s new AI chatbot will help you build virtual worlds
/ Roblox’s new AI assistant is one of a few new AI tools from the company.
https://www.theverge.com/2023/9/8/23863943/roblox-ai-chatbot-assistant-ai-rdc-2023
Tomi Engdahl says:
Alan Rappeport / New York Times:
The IRS starts using AI to investigate complex cases of tax evasion at multibillion-dollar partnerships like hedge funds, private equity groups, and law firms
https://www.nytimes.com/2023/09/08/us/politics/irs-deploys-artificial-intelligence-to-target-rich-partnerships.html
Tomi Engdahl says:
https://hackaday.com/2023/09/07/programming-a-poker-game-with-gpt-help/
Tomi Engdahl says:
Mushroom pickers urged to avoid foraging books on Amazon that appear to be written by AI
https://www.theguardian.com/technology/2023/sep/01/mushroom-pickers-urged-to-avoid-foraging-books-on-amazon-that-appear-to-be-written-by-ai
Sample of books scored 100% on AI detection test as experts warn they contain dangerous advice
Amateur mushroom pickers have been urged to avoid foraging books sold on Amazon that appear to have been written by artificial intelligence chatbots.
Amazon has become a marketplace for AI-produced tomes that are being passed off as having been written by humans, with travel books among the popular categories for fake work.
Now a number of books have appeared on the online retailer’s site offering guides to wild mushroom foraging that also seem to be written by chatbots. The titles include “Wild Mushroom Cookbook: form [sic] forest to gourmet plate, a complete guide to wild mushroom cookery” and “The Supreme Mushrooms Books Field Guide of the South-West”.
Four samples from the books were examined for the Guardian by Originality.ai, a US firm that detects AI content. The company said every sample had a rating of 100% on its AI detection score, meaning that its systems are highly confident that the books were written by a chatbot such as ChatGPT.
Tomi Engdahl says:
Generative AI Generation Gap: 70% Of Gen Z Use It While Gen X, Boomers Don’t Get It
https://www.forbes.com/sites/johnkoetsier/2023/09/09/generative-ai-generation-gap-70-of-gen-z-use-it-while-gen-x-boomers-dont-get-it/?fbclid=IwAR0u-DAnD6ATOrnoHSAF_nBnU4pcyM-oWhQwDa5Q3E-myXtk7d2xgcQOUXg&sh=66c491c923b6
75% of people who use generative AI use it for work and 70% of Gen Z uses new generative AI technologies, according to a new 4,000-person survey by Salesforce, which has been integrating AI into its products for years. Also found: 68% of those who haven’t tried generative AI are Gen X or boomers.
“There’s a generative AI divide,” says Salesforce senior director of product marketing Kelly Eliyahu. “49% of the population has used it, and 51% has never used it.”
Tomi Engdahl says:
Financial Times:
An in-depth look at how Google’s Transformer model, developed by eight researchers in 2017, radically sped up and augmented how computers understand language — Over the past few years, we have taken a gigantic leap forward in our decades-long quest to build intelligent machines: the advent of the large language model, or LLM.
https://subs.ft.com/products?location=https%3A%2F%2F%2Fgenerative-ai%2F
Tomi Engdahl says:
Steven Levy / Wired:
Q&A with Sundar Pichai on 25 years of Google, competition in AI, ChatGPT in Bing, DeepMind and Google Brain, AI chips, staff’s bureaucracy complaints, and more — The tech giant is 25 years old. In a chatbot war. On trial for antitrust. But its CEO says Google is good for 25 more.
https://www.wired.com/story/sundar-pichai-google-ai-microsoft-openai/
Tomi Engdahl says:
Dylan Martin / CRN:
Nvidia claims TensorRT-LLM will double the H100′s performance for running inference on leading LLMs when the open-source library arrives in NeMo in October
Nvidia Says New Software Will Double LLM Inference Speed On H100 GPU
Dylan Martin
September 08, 2023, 02:42 PM EDT
https://www.crn.com/news/components-peripherals/nvidia-says-new-software-will-double-llm-inference-speed-on-h100-gpu
The AI chip giant says the open-source software library, TensorRT-LLM, will double the H100’s performance for running inference on leading large language models when it comes out next month. Nvidia plans to integrate the software, which is available in early access, into its Nvidia NeMo framework as part of the Nvidia AI Enterprise software suite.
Tomi Engdahl says:
Blake Brittain / Reuters:
The US Copyright Office denies protection for an AI-made image that won an art competition, despite Adobe Photoshop alterations and 624+ text prompt revisions
US Copyright Office denies protection for another AI-created image
https://www.reuters.com/legal/litigation/us-copyright-office-denies-protection-another-ai-created-image-2023-09-06/
Summary
Midjourney-created “space opera” artwork not protectable, office says
Office has previously rejected copyrights for AI-generated work
Sept 6 (Reuters) – The U.S. Copyright Office has again rejected copyright protection for art created using artificial intelligence, denying a request by artist Jason M. Allen for a copyright covering an award-winning image he created with the generative AI system Midjourney.
The office said on Tuesday that Allen’s science-fiction themed image “Theatre D’opera Spatial” was not entitled to copyright protection because it was not the product of human authorship.