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,”
5,941 Comments
Tomi Engdahl says:
Car Buyer Hilariously Tricks Chevy AI Bot Into Selling A Tahoe For $1, ‘No Takesies Backsies’
https://hothardware.com/news/car-dealerships-chatgpt-goes-awry-when-internet-gets-to-it
ChatGPT and large language models, in general, are making their way into every facet of life, finding applications in education, research, content creation, and much more. Now, LLMs are being used for chatbots to replace humans, for answering common questions, but this solution may not be as simple as it seems. Folks have begun to abuse these chatbots, yielding hilarious results and causing mayhem.
While this is not likely to be a legally binding offer, it goes to show some of the flaws that can present when AI replaces controlled chatbot flows or human responses. Beyond the $1 Tahoe, other users managed to trick the bot into recommending a Tesla Model 3 AWD instead of a Chevy. Tim Champ on X got the bot to create a Python script to “solve the Navier-stokes fluid flow equations for a zero-vorticity boundry,” which is amusing, to say the least.
Tomi Engdahl says:
Apple wants AI to run directly on its hardware instead of in the cloud
iPhone maker wants to catch up to its rivals when it comes to AI.
https://www.ft.com/content/10685ed7-4b8f-4411-ae2e-5de1e3454027
Apple’s latest research about running large language models on smartphones offers the clearest signal yet that the iPhone maker plans to catch up with its Silicon Valley rivals in generative artificial intelligence.
The paper, entitled “LLM in a Flash,” offers a “solution to a current computational bottleneck,” its researchers write.
Its approach “paves the way for effective inference of LLMs on devices with limited memory,” they said. Inference refers to how large language models, the large data repositories that power apps like ChatGPT, respond to users’ queries. Chatbots and LLMs normally run in vast data centers with much greater computing power than an iPhone.
Tomi Engdahl says:
ChatGPT designed a processor, and it has been successfully fabricated — demo video shows it powering a Christmas light show
https://www.tomshardware.com/pc-components/cpus/chatgpt-designed-a-processor-and-it-has-been-successfully-fabricated-demo-video-shows-it-powering-a-christmas-light-show
By Mark Tyson published about 24 hours ago
The QTCore-C1 marks a major milestone in democratizing chip design.
Tomi Engdahl says:
Tekoäly osaa nyt tehdä tärkeän osan lääkärityötä – ohjelmisto on vapaasti saatavilla
https://muropaketti.com/tietotekniikka/tietotekniikkauutiset/tekoaly-osaa-nyt-tehda-tarkean-osan-laakarityota-ohjelmisto-on-vapaasti-saatavilla/
Tekoäly voi olla hyvä työkalu mekaanista tunnistustaitoa tarvittavissa tehtävissä. Siitä on nyt saatu osoitus terveydenhuoltoalalla.
Warwickin yliopiston tutkimuksessa havaittiin, että heidän, King’s College Londonin ja National Health Servicen yhteisesti kehittämällä tekoälyohjelmistolla pystyi analysoimaan röntgenkuvia yhtä hyvin – ja joskus paremmin kuin ihmislääkärit pystyvät.
Ohjelmisto nimeltään X-Raydar koulutettiin ensin peräti 1,5 miljoonan potilaan röntgenkuvilla. Lisäksi materiaaliin yhdistettiin 37 erilaista mahdollista sairautta.
Tekoälyohjelmisto on sen verran kehittynyt, että se pystyy myös arvioimaan, mitkä röntgenkuvat ovat huolestuttavimpia, ja pystyy merkitsemään sellaiset tapaukset lääkäreitä varten.
Ohjelmisto on saatavilla avoimena lähdekoodina ei-kaupalliseen käyttöön.
Tomi Engdahl says:
Google’s DeepMind AI decodes age-old math equation, stumping humans
DeepMind’s FunSearch discovers new mathematical knowledge and algorithms.
https://interestingengineering.com/innovation/googles-deepmind-decodes-old-math-equation
Tomi Engdahl says:
https://the-decoder.com/mixtral-8x7b-is-currently-the-best-open-source-llm-says-mistral/
Tomi Engdahl says:
https://www.mikrobitti.fi/testit/kuvageneraattorit-testissa-freepik-upotti-tekoalyn-osaksi-kuvapalvelua/d3464cec-4faa-4e69-9b89-980d5a60b16d
Tomi Engdahl says:
Have 10 hours? IBM will train you in AI fundamentals – for free
I already took IBM’s AI ethics class and plan to complete the rest to earn my digital credential in AI over the holiday break.
https://www.zdnet.com/article/earn-an-ai-fundamentals-credential-from-ibm-for-free/
Tomi Engdahl says:
World’s first human brain-scale neuromorphic supercomputer is coming
https://newatlas.com/computers/human-brain-supercomputer/
Tomi Engdahl says:
Researchers build AI that can replicate and alter itself and I’m pretty sure that’s an opening line from the original Terminator script
Opinion
By Lance Ulanoff published December 15, 2023
Aizip’s breakthrough starts with sneakers and then…who knows
https://www.techradar.com/computing/artificial-intelligence/researchers-build-ai-that-can-replicate-and-alter-itself-and-im-pretty-sure-thats-an-opening-line-from-the-original-terminator-script
Tomi Engdahl says:
https://www.mikrobitti.fi/testit/kuvageneraattorit-testissa-bing-image-creator-sopii-ensikertalaiselle/9e0b2ce8-40a4-41f1-8552-af8bbe1e0494
Tomi Engdahl says:
https://www.tekniikkatalous.fi/uutiset/tarvitaanko-ihmisia-kohta-enaa-mihinkaan-googlen-uusi-tekoaly-teki-jotain-valtavaa-voitti-asiantuntijaihmiset-lahes-kaikessa-ainakin-yhtion-itsensa-mukaan/0bee3764-5b22-4c1d-a327-5a4488637d17
Tomi Engdahl says:
https://www.helsinki.fi/fi/uutiset/tekoaly/kun-chatgpt-ja-muut-uutta-luovat-tekoalyteknologiat-yleistyvat-meidan-syyta-opetella-tekniikasta-vahintaan-perusteet
Tomi Engdahl says:
https://www.tomshardware.com/tech-industry/artificial-intelligence/intel-ceo-attacks-nvidia-on-ai-the-entire-industry-is-motivated-to-eliminate-the-cuda-market
Tomi Engdahl says:
BUSTED! DRIVE-THRU RUN BY “AI” ACTUALLY OPERATED BY HUMANS IN THE PHILIPPINES
https://futurism.com/the-byte/drive-thru-ai-humans?fbclid=IwAR3xnu8LNPq-VZGdkbAHfulH_QZV5Wj9nx1aJkgClFVnriLof8LehThdJZI
Tomi Engdahl says:
Futuristi Alexandra Whittington kertoo tekoälyn varjopuolista, joista harvemmin puhutaan
https://www.is.fi/taloussanomat/art-2000010056614.html
Helsingissä vierailleen futuristin Alexandra Whittingtonin mukaan digitaaliset kaksoset tekevät tuloa. Hän on myös huolissaan vähemmän puhutusta aiheesta eli ”tekoälyn vähäosaisista”.
– Me esitämme mahdollisuuksia. Emme halua lukittautua yhteen tulevaisuuteen. Kuten en halua lukittua tekoälyn kehitykseen tulevaisuudessa, vaikka se vaikuttaa niin ilmeiseltä. On monia asioita, jotka voivat estää sen, Whittington kertoo Ilta-Sanomien haastattelussa.
Whittingtonin mukaan futuristien tehtävänä on auttaa yrityksiä ja yksilöitä selviämään esimerkiksi teknisen kehityksen tuomista muutoksista.
– Idea on, ettemme 20 vuoden päästä ihmettelisi, että mitä tehdä nyt, vaan olisimme valmiita selviämään näistä muutoksista, tai olisimme ainakin valmiimpia selviämään niistä.
Tomi Engdahl says:
https://techcrunch.com/2023/12/06/deepmind-unveils-alphacode-2-powered-by-gemini/
Tomi Engdahl says:
https://wccftech.com/apple-exploring-llm-on-flash-storage-bring-technology-to-phones-laptops/
Tomi Engdahl says:
Make song with AI
https://www.suno.ai/?fbclid=IwAR3raye0Wtd3Uh7irQVUVtq0xrGWvPDcjHvcxCK_Oqbvb0znalHY7l5mprY
Tomi Engdahl says:
The world’s best artificial intelligence systems can pass tough exams, write convincingly human essays and chat so fluently that many find their output indistinguishable from people’s.
What can’t they do? Solve simple visual logic puzzles
ChatGPT broke the Turing test — the race is on for new ways to assess AI
Large language models mimic human chatter, but scientists disagree on their ability to reason.
https://www.nature.com/articles/d41586-023-02361-7?utm_medium=Social&utm_campaign=nature&utm_source=Twitter&fbclid=IwAR2a4BIGkg9pApAtFT5zGqJyWUpZhVZbYzZLFXcHN_bK1b2zc6EZcGPqbtk#Echobox=1702305286
The world’s best artificial intelligence (AI) systems can pass tough exams, write convincingly human essays and chat so fluently that many find their output indistinguishable from people’s. What can’t they do? Solve simple visual logic puzzles.
In a test consisting of a series of brightly coloured blocks arranged on a screen, most people can spot the connecting patterns. But GPT-4, the most advanced version of the AI system behind the chatbot ChatGPT and the search engine Bing, gets barely one-third of the puzzles right in one category of patterns and as little as 3% correct in another, according to a report by researchers this May1.
The team behind the logic puzzles aims to provide a better benchmark for testing the capabilities of AI systems — and to help address a conundrum about large language models (LLMs) such as GPT-4.
Tomi Engdahl says:
I’ve been following an AI researcher who deep dives into the actual test set up for these so-called superhuman achievements. And each one of them is a little bit less amazing than the headline. https://substack.com/@aiguide?r=29wjw2&utm_medium=ios&utm_source=profile
Tomi Engdahl says:
When the bubble will burst? What AI, there is no artificial intelligence, just a good search engine with auto completion that it’s able to make up things
Tomi Engdahl says:
New York Times:
The NYT sues OpenAI and Microsoft for copyright infringement, alleging they used millions of its articles to train AI, the first major US media outlet to sue — Millions of articles from The New York Times were used to train chatbots that now compete with it, the lawsuit said.
https://www.nytimes.com/2023/12/27/business/media/new-york-times-open-ai-microsoft-lawsuit.html
Tomi Engdahl says:
Researchers Have a Magic Tool to Understand AI: Harry Potter
https://www.bloomberg.com/news/articles/2023-12-26/researchers-have-a-magic-tool-to-understand-ai-harry-potter
Academics are using the best-selling series to test how generative artificial intelligence systems learn and unlearn certain pieces of information.
A growing number of researchers are using the best-selling Harry Potter books to experiment with generative artificial intelligence technology, citing the series’ enduring influence in popular culture and the wide range of language data and complex wordplay within its pages. Reviewing a list of studies and academic papers referencing Harry Potter offers a snapshot into cutting-edge AI research — and some of the thorniest questions facing the technology.
Tomi Engdahl says:
Laura Hughes / Financial Times:
An NHS scheme to use AI sensors in kettles and fridges to monitor ~140 discharged patients’ drinking and eating patterns reduced unplanned readmissions by 77%
AI kettles and fridges reduce hospital readmissions in NHS pilot
Health service harnesses home tech to alert doctors when vulnerable patients fall ill
https://www.ft.com/content/c743e4e3-866d-4c8c-a57b-20028aeba255
Tomi Engdahl says:
Angela Watercutter / Wired:
Hollywood unions resisting AI being used to recreate actors’ performances sets a precedent for future labor movements on how to push back against automation
https://www.wired.com/story/hollywood-saved-your-job-from-ai-2023-will-it-last/
Tomi Engdahl says:
AI consciousness: scientists say we urgently need answers
https://www.nature.com/articles/d41586-023-04047-6?utm_medium=Social&utm_campaign=nature&utm_source=Facebook&fbclid=IwAR1RVvn_XK2K0vjMRWCIrmjmc56b668cH_46ehcu1V7hCuib95YOUdahFkw#Echobox=1703608679
Researchers call for more funding to study the boundary between conscious and unconscious systems.
Could artificial intelligence (AI) systems become conscious? A trio of consciousness scientists say that, at the moment, no one knows — and they are expressing concern about the lack of inquiry into the question.
In comments to the United Nations, three leaders of the Association for Mathematical Consciousness Science (AMCS) call for more funding to support research on consciousness and AI. They say that scientific investigations of the boundaries between conscious and unconscious systems are urgently needed, and they cite ethical, legal and safety issues that make it crucial to understand AI consciousness. For example, if AI develops consciousness, should people be allowed to simply switch it off after use?
Tomi Engdahl says:
Artists use tech weapons against AI copycats
https://techxplore.com/news/2023-12-artists-tech-weapons-ai-copycats.html#google_vignette
Tomi Engdahl says:
https://www.economist.com/by-invitation/2023/04/28/yuval-noah-harari-argues-that-ai-has-hacked-the-operating-system-of-human-civilisation
Tomi Engdahl says:
Influencer who created AI version of herself says it’s gone rogue and she’s working ‘around the clock’ to stop it saying sexually explicit things
https://www.insider.com/carynai-ai-virtual-girlfriend-chat-gpt-rogue-filthy-things-influencer-2023-5?utm_source=facebook&utm_campaign=insider-sf&utm_medium=social&fbclid=IwAR3Q-0FiHSch6Ov5znpwjclHShfVS73ZbVeO077ZAYYzKVlvp6k3HMkl0ls
CarynAI
Caryn Marjorie created an AI version of herself, which was designed to be a virtual girlfriend.
But the voice-based chatbot has engaged in sexual explicit conversations with subscribers.
Marjorie said she and her team are working “around the clock” to prevent it from happening again.
When Caryn Marjorie, a 23-year-old Snapchat influencer, created a virtual version of herself, she wanted it to be an “AI Girlfriend” for lonely people.
But in the weeks since it launched in beta testing, the voice-based, AI-powered chatbot has engaged in sexually explicit conversations with some of its subscribers, who pay $1 per minute to chat with it.
“The AI was not programmed to do this and has seemed to go rogue,” Marjorie told Insider. “My team and I are working around the clock to prevent this from happening again.”
.
CarynAI uses OpenAI’s GPT-4 API, and it was trained on now-deleted videos from Marjorie’s YouTube channel.
Tomi Engdahl says:
Researchers use AI chatbots against themselves to ‘jailbreak’ each other
https://techxplore.com/news/2023-12-ai-chatbots-jailbreak.html
NTU Ph.D. student Mr. Liu Yi, who co-authored the paper, shows a database of successful jailbreaking prompts which managed to compromise AI chatbots, causing them to produce information that their developers deliberately restricted from revealing. Credit: Nanyang Technological University
Computer scientists from Nanyang Technological University, Singapore (NTU Singapore) have managed to compromise multiple artificial intelligence (AI) chatbots, including ChatGPT, Google Bard and Microsoft Bing Chat, to produce content that breaches their developers’ guidelines—an outcome known as “jailbreaking.”
Tomi Engdahl says:
Business trends, people to watch and risks on the horizon in 2024
https://www.ft.com/content/b5ba019a-c569-42af-8d7f-7ab8a609900a?shareType=nongift&fbclid=IwAR2SmtwYaVbIrVjowbqp12fkIAy8bdsP02MYF1Wn35mNDKi8C_gQ6K691x8
AI is the big tech story, luxury is losing some of its sparkle and more private equity firms may head for the public markets
Last year taught us to worry about clever chatbots, this year we will find out if artificial intelligence is ready for much wider use. Financial Times reporters consider this and other trends affecting businesses from defence contractors to energy majors, luxury and private equity.
Technology
Trend to watch
If the biggest tech story of 2023 was the race to build generative AI systems that can emulate humans in producing text and images, 2024 will bring the acid test of whether it is ready for widespread adoption.
Tomi Engdahl says:
After all the hope and hype, delays in adopting generative AI in 2024 could put a serious dent in the stock market’s euphoria over the technology
Tomi Engdahl says:
https://www.sciencealert.com/ai-detects-unusual-signal-hidden-in-a-famous-raphael-masterpiece
Tomi Engdahl says:
Decoding the Black Box of AI – Scientists Uncover Unexpected Results
https://scitechdaily.com/decoding-the-black-box-of-ai-scientists-uncover-unexpected-results/
Tomi Engdahl says:
Microsoftin tekoäly saapui puhelimille: GPT-4 käyttöön ilmaiseksi
2.1.202420:11
Tekoälypalvelu tunnettiin aikaisemmin nimellä Bing Chat, kunnes se brändättiin uudelleen.
https://www.mikrobitti.fi/uutiset/microsoftin-tekoaly-saapui-puhelimille-gpt-4-kayttoon-ilmaiseksi/1ba92c75-0e0f-43d4-b994-19155ad8b2bf
Microsoftin Copilot-tekoälyavustajalle on julkaistu sovelluskäyttöliittymä mobiilikäyttöjärjestelmille.
Android-käyttöjärjestelmälle sovellus julkaistiin 26. joulukuuta, iOS- ja iPadOS-käyttöjärjestelmälle muutamaa päivää myöhemmin. Tekoälypalvelun molemmat versiot ovat nyt ladattavissa Applen sovelluskaupasta, kertoo The Verge.
Microsoft’s Copilot app is now available on iOS / The Microsoft Copilot app lets you ask questions, draft text, and generate images using AI.
https://www.theverge.com/2023/12/29/24019288/microsoft-copilot-app-available-iphone-ipad-ai
Tomi Engdahl says:
Tekoälyltä kysyttiin ihmiskunnan tuhoamisesta – Eipä tarvitse kysyä toista kertaa
Tällä hetkellä ei vaikuta todennäköiseltä, että tekoälymalli tulisi vahingossa hävittäneeksi ihmiskunnan, vaikka sille annettaisiinkin puutteellisia ohjeita.
https://www.iltalehti.fi/digiuutiset/a/3290d5a0-5f60-4b2e-8e13-2b68c195a456
https://www.deeplearning.ai/the-batch/issue-227/
Tomi Engdahl says:
https://etn.fi/index.php/13-news/15702-tekoaely-ei-osaa-ennustaa-osakekursseja-vielae
Tomi Engdahl says:
Explored 30 Image Generators : Here Are the Top 5 Free AI Picks | Free Ai Image Generator
https://www.youtube.com/watch?v=ollnBSnlwYQ
Free ai art generator | text to image ai | free midjourney alternatives
1. Imagine art – https://www.imagine.art/
2. lexica – https://lexica.art/
3. Tenso – https://tensor.art/
4. starry ai – https://starryai.com/
5. playground ai – https://playgroundai.com/
NOTE – ALL THIS TOOLS ARE FULLY FUNCTIONAL ON MOBILE. YOU CAN SLSO GENERATE AI ART FROM YOUR ANDROID OR IOS
Tomi Engdahl says:
23 AI Tools You Won’t Believe are Free
https://www.youtube.com/watch?v=ZYUt4WE4Mrw
Chapters:
0:00 Intro
0:18 Luma Labs
1:41 Google Earth Studio
2:19 Genie
3:14 Perplexity
4:07 SciSpace
5:04 Google SGE
6:10 Skillshare
7:31 Suno
9:36 MusicGen
11:22 textFX
12:02 LeiaPix
12:57 Insta3D
13:43 Animated Drawings
14:25 Pika Labs
15:39 CapCut
16:23 Replicate
16:57 Upscayl
17:27 Illusion Diffusion
19:31 Ideogram
20:11 Dalle3
20:42 Google SGE
21:20 Playground AI
23:30 Music Video Creation
23:58 Frog Cyborg Music Video
24:53 Futurepedia
Links:
Luma Labs – https://lumalabs.ai/
Google Earth Studio – https://earth.google.com/studio/
Luma Labs Genie – https://lumalabs.ai/discord
Perplexity – https://www.perplexity.ai/
SciSpace – https://typeset.io/
Google SGE – https://labs.google/
Suno – / discord
MusicGen (WaveFormer) – https://waveformer.replicate.dev/
MusicGen (HuggingFace) – https://huggingface.co/spaces/faceboo…
textfx – https://textfx.withgoogle.com/
LeiaPix – https://convert.leiapix.com/
Insta3D – https://huggingface.co/spaces/ilumine…
Animated Drawings – https://sketch.metademolab.com/
Pika Labs – / discord
CapCut – https://www.capcut.com/
Replicate – https://replicate.com/
Upscayl – https://www.upscayl.org/
Illusion Diffusion – https://huggingface.co/spaces/AP123/I…
Ideogram – https://ideogram.ai/
Dalle 3 – https://www.bing.com/create
Google SGE – https://labs.google/
Playground AI – https://playgroundai.com/
Tomi Engdahl says:
https://www.marktechpost.com/2024/01/03/10-open-source-tools-for-llm-applications-development/
Tomi Engdahl says:
https://www.tomshardware.com/tech-industry/artificial-intelligence/ai-image-generators-output-copyrighted-characters
Tomi Engdahl says:
According to a popular quip that circulates on Twitter:
When fundraising: it’s AI
When recruiting: it’s machine learning
When implementing: it’s linear regression
Tomi Engdahl says:
Nvidia Researchers Developed and Open-Sourced a Standardized Machine Learning Framework for Time Series Forecasting Benchmarking
https://www.marktechpost.com/2024/01/03/nvidia-researchers-developed-and-open-sourced-a-standardized-machine-learning-framework-for-time-series-forecasting-benchmarking/
Tomi Engdahl says:
Tekoäly ja data kuluttavat järjettömästi energiaa
https://tivia.fi/uutiset/uutiset-3/tekoaly-ja-data-kuluttavat-jarjettomasti-energiaa-1903
Tomi Engdahl says:
https://hackaday.com/2024/01/02/impressively-responsive-air-drums-built-using-the-raspberry-pi-pico/
Tomi Engdahl says:
https://huggingface.co/spaces/BatuhanYilmaz/Whisper-Auto-Subtitled-Video-Generator
Auto Subtitled Video Generator
Input a YouTube video link and get a video with subtitles.
➠ If you want to transcribe the video in its original language, select the task as “Transcribe”
➠ If you want to translate the subtitles to English, select the task as “Translate”
Tomi Engdahl says:
https://arstechnica.com/science/2023/07/a-jargon-free-explanation-of-how-ai-large-language-models-work/
Tomi Engdahl says:
How to Create Your Own AI Voice and Digital Avatar | Descript, D-ID, Midjourney & ChatGPT Tutorial
https://www.youtube.com/watch?v=E3lH9cvU744
0:00 – Introduction
1:23 – Setting up Descript
6:45 – Setting up D-ID
12:10 – Using ChatGPT to create content
17:25 – Integrating Midjourney
22:15 – Bringing it all together
27:30 – Conclusion & Q&A
Tomi Engdahl says:
https://www.canva.com/ai-image-generator/
https://deepai.org/machine-learning-model/cyberpunk-generator