Hi, AI! | media

🦆 AI Saves Migratory Birds

Wind is an important source of renewable energy in the US and Europe. The EU plans to increase its use sixfold by 2030. However, offshore wind turbines cause an estimated 4.5 million bird deaths each year. Raptor birds usually look downwards rather than forwards when flying in search of prey and therefore often get caught in the turbines’ spinning blades. The death of birds, in turn, damages local ecosystems and agriculture.

Norwegian startup Spoor, which recently raised a $4 million seed round from investors, created an artificial intelligence-based technology that will help save hundreds of thousands of birds.

💡 How It Works

Cameras on wind turbines detect birds around the clock. The AI-powered bird-tracking technology counts them, analyzes their trajectories, and estimates the risks of collisions with the blades using 3D models.

The birds fly at a speed of about 10 m/s. Spoor software can detect and track birds up to 2 km away using video. The turbine receives a signal and can slow down to two revolutions per minute in 20-40 seconds, becoming safe for the birds.

Wind farms are quite huge, many hundreds of square kilometers, and trying to use computer vision to basically monitor the air is an interesting technology challenge. We needed to create a scalable technology that can detect birds. It’s kind of a novel use of computer vision and our own data pipeline.

Ask Helseth, co-founder and CEO of Spoor

👍 Why It Is Useful?

👍 Hundreds of thousands of birds will be saved!

👍 The bird migration dataset can help wind farms slow down or even stop turbines in real-time when avian activity is expected to increase.

👍 Companies will be able to find and monitor safer places to build wind farms by assessing risks to the local avian populations, especially endangered bird species.

#startup @hiaimediaen

📣 Hello! Welcome to our Sunday digest, featuring the most exciting AI news from the 24th week of 2024.

MAIN NEWS

📱 Apple Introduced AI System for Its Devices — Apple Intelligence. Users can expect a new Siri, personalized AI features, and a built-in free ChatGPT.

SAVE THIS — IT'S HELPFUL

▶️ Luma Dream Machine — Next Generation AI Video. This model creates five-second clips based on simple text and image inputs.

✈️ Around the World with ChatGPT. Sharing our favorite prompts for planning any trip.

💬 Character AI — a Conversation Partner on Any Topic. One of the most popular AI-powered apps lets you chat with any character or design your own.

TO READ

📱 How AI is Changing the Film Industry: Insights from a Netflix CEO, AI-Generated series, and critiques of The Last of Us creator.

🎨 AI Reads Your Thoughts and Creates Paintings. The Obvious mind-to-image project allows you to visualize your imagination.

🆒 How Generative AI is Changing the World. World-renowned leader in quantum computing and IBM Vice President Dario Gil shared insights in an interview with The New Yorker.

🔒 Five Challenges to Achieve AGI: Overcoming these problems could change the field and help us create Artificial General Intelligence.

TO WATCH

📱 AI Generation: A 14-year-old tech enthusiast gave a TEDx talk titled "From AI Curious to AI Serious."

Have a great end of the week 🍀

#AIweek | @hiaimediaen

🐰 AI Generation: 14-Year-Old Tech Enthusiast Gives TEDx Talk on AI

Raul John, known as #Raultherockstar, is a rising star on @Techonaroll' rel='nofollow'>YouTube and Instagram, covering technology, AI, and IT education.

🆒 A Young Prodigy from India

Raul was born and raised in Edappally, a provincial town in India. In 2018, he launched his YouTube channel, where he engagingly discusses technology and creates content for his peers. Last year, he launched online AI courses for American schoolchildren. In his lessons, Raul is accompanied by his homemade robot, MeBot.

The robot's core includes several AI engines: Canva AI, Unreal Engine, and Inworld AI. But the most amazing thing is that Raul and his classmate assembled the robot himself as an exhibit for a school science fair (of course, he won the grand prize!). According to this AI wunderkind, he created MeBot to answer questions from American schoolchildren when he sleeps. The robot speaks in Raul's voice 🇨🇦

"My goal is to use technology to improve the future. I create videos daily about how technology can positively impact our lives. And I talk about AI tools for content creators, video editors, and even lawyers," — Raul John, teenager, and AI influencer

🎤 What Was His Lecture About?

Raul gave an interesting talk on the topic "From AI Curious to AI Serious." Here is a summary of his lecture:

🔣 When Raul started his YouTube channel, it took him about three months to edit one video. He found an editing program that correctly cut frames, remove pauses, do zooms, and smooth transitions. Raul studied the program and learned that AI did all this for him, which led him to start sharing the benefits of this technology on the Clubhouse platform. Gradually, his audience grew, people began gifting Raul subscriptions to various AI models for testing, and his opinion became authoritative.

🔣 Raul doubts that modern AI can be called "intelligent." As evidence, he discusses the "Chinese Room" experiment, which refutes the possibility of AI possessing consciousness (of course, this does not refer to AGI).

🔣 Why has AI become popular now, not 15-20 years ago? The main role was GPUs (graphics processors), which can perform repetitive operations and handle multiple tasks simultaneously, unlike CPUs (central processors, which are like the CEO of a company — important but slow and capable of solving only one task at a time). The mass production and continuous performance improvement of GPUs have made AI models so effective.

More stories:

📹 5 Engaging TED Talks on AI

#ted @hiaimediaen

📱 AI Instead of Screenwriters: Forecasts from Netflix CEO, New AI Streaming Series, and The Last of Us

Ted Sarandos, one of the executives at Netflix, is not concerned about the implementation of AI in the film industry. In an interview with the New York Times, he called AI a natural stage in the evolution of the creative industry and expressed doubt that AI would take away jobs from screenwriters and directors.

Two key points from the interview:

1️⃣ A Tool for Increased Efficiency. Screenwriters, directors, and editors will use AI as a tool to work more efficiently and effectively. Just as computer animation did not replace hand-drawn animation, AI will help create what was previously impossible. Every new achievement in the industry leads to its development.

2️⃣ AI Won't Take Your Job. It won't be able to write a script better than a great writer. AI won't take your job, but a person who knows how to use AI might.

⭐ AI-Netflix

Last week, San Francisco startup Fable Studio launched an AI streaming platform called Showrunner. All content on it is created by generative AI, and users can create new episodes themselves. The service already has a waiting list of 50,000 people. You can watch a free episode of the animated series "Exit Valley," which humorously portrays life in Silicon Valley, similar to the style of "South Park."

🍄 The Last of Us Scandal

"The Last of Us," an HBO series based on the hit video game, has received rave reviews from both critics and fans. Neil Druckmann, the main developer of the game, recently sparked controversy after discussing the role of AI in video game storytelling during an interview:

1️⃣ Thanks to AI, we are reducing costs while simultaneously opening up the possibility of tackling bolder projects and pushing the boundaries of storytelling in games.

2️⃣ AI will allow for more realistic dialogues and characters. However, it is important to use these tools purposefully to achieve specific goals.

The praise for AI as a game scriptwriter sparked criticism from fans of The Last of Us and Druckmann's colleagues. Many commentators emphasized that the success of The Last of Us is due to the work of screenwriters, not AI. Shortly after the interview, Sony removed the conversation, citing errors, and Druckmann apologized, stating that his words had been misunderstood.

More stories:

Strikes by Hollywood Screenwriters Against AI

Winners Announced for AI Film Festival 2024

AI in the Movie "Furiosa"

#netflix @hiaimediaen

🆒 World-renowned leader in quantum computing and IBM Vice President Dario Gil recently shared in an interview with The New Yorker how generative AI is changing the world.

🥸 Who is Dario Gil?

⚫️Director of IBM Research, responsible for the company's scientific developments, including quantum computing and AI.
⚫️Manages an annual budget of $6.9 billion.
⚫️Oversees a team of over 3,000 researchers, including 6 Nobel laureates.
⚫️Under his leadership, IBM created the first programmable quantum computers and made them accessible via the cloud.

Interesting insights from the interview:

🔼 A Turning Point in History. We're at a significant inflection point, and it feels like the equivalent of the first browsers when they appear and people realize the potential of the Internet. In the past nine months, the world has experienced something similar.

➕ On Future Exams. With the advent of pocket calculators, professors changed exams by removing calculation tasks. The same will happen with AI. Automatable tasks will disappear, leaving deeper and more creative questions. Written exams without computers may return.

🤖 On Understanding the World with AI. We will use a hybrid way of understanding the world with AI. AI will provide answers, often without explanations. The task of scientists in the future will be to find these explanations.

❓ A New Paradigm for the World. AI will become more accessible to many, increasing people's efficiency. However, there will be fewer capable of creating AI. This could increase inequality between countries and companies, between the rich and the poor. It's important not only to use AI but also to be able to be an AI value creator.

⚠️ Should Actors be Protected from AI? Yes, they should. Consider voice cloning. Producers can film an actor for five minutes and then generate hours of content, paying only for a short session. This reduces the income of actors who are paid by the hour.

⚙️ What Problems Hinder Companies from Implementing AI?

▶ Business processes. Implementing AI requires global changes in workflows and relationships between people.
▶ Data. Companies often do not understand how to process and create data for AI models properly.
▶ Resources. Effective use of AI requires large teams and significant financial investments in computing.

More stories:

🤖 Yann LeCun, Meta — on AI Dominance, OpenAI's Secrecy, and Superintelligence

#ibm @hiaimediaen

💬 Character.AI — is one of the most popular AI-powered apps in the world. This AI chatbot service allows you to communicate with any fictional character or design your own. More than 3.5 million people use the platform daily.
Now Character is valued at over $1 billion. Over 60% of users are aged 18-24 years.

More than 18 million AI-powered bots have already been created on the platform. Each has its own “character.” These include alter egos of famous people from Adam to Jean-Claude van Damme, fictional personas from books, fantasy, anime, and video games, as well as practical bots such as career coaches, psychotherapists, and language teachers.

👀 The Verge recently published an investigation about why millions of teenagers around the world spend a lot of time on the app, often preferring AI companions to real-life friends. Users communicate with bots by assigning them certain personalities, interests, and communication styles. For many, it's an anonymous and safe space to express themselves. Still, there is a warning above every chat room: “Everything Characters say is made up!”

Psychotherapy and relationships are among the most visited topics. The most popular Psychologist bot has had around 115 million chats at the time of writing.

“Research shows that chatbots can aid in lessening feelings of depression, anxiety, and even stress. However, it’s important to note that many of these chatbots have not been around for long periods of time and are limited in what they can do,” —
Kelly Merrill Jr., professor at the University of Cincinnati.


ℹ️ How does it work?

After downloading the free app, you need to select a bot in search or recommendations and start a conversation — either in writing or voice. The voiceover is available in eight languages.

📱 How are the bots trained?

The model is trained on real conversations from millions of Reddit, X, and other platforms’ users. The bot adapts to the user’s communication style: it mirrors their manner of speech, and sense of humor, and remembers the context of the conversation.

More on the topic:

✌️ Character.AI is the second most visited AI service

#startup @hiaimediaen

🔒 What We Lack on the Path to AGI

While CEOs of various companies argue about what AGI really is, developers have already compiled a list of problems facing AI. Solving these issues might revolutionize the field and lead us to AGI.

💬 AGI vs. AI

One definition of AGI is the ability of AI to perform most tasks that a human can do well. Although there are models today that successfully create texts or images, they are still far from being universal.

☄️ 5 Steps to AGI That Will Strengthen AI

1. Acquiring different ways of thinking (deduction, association) and memory.
2. AI should be able to learn and conduct experiments.
3. Natural language processing is necessary for interaction with humans.
4. The AI agent should gather information from reality.
5. Consciousness — the most controversial point, as even in the case of humans, it is unclear how consciously they perform actions.

🤯 What Challenges Does AI Currently Face?

⚫️ Huge Data Sets

Data sets are becoming large and complex, making traditional information processing methods ineffective. Developers propose dividing data sets into easily manageable segments and implementing RAG, a method of advanced information retrieval. This method combines LLM and a search system across all the data you have.

⚫️ Deductive Logic

The main problem with deep learning is generalized rigid reasoning; LLMs are still not capable of making inferences based on provided information. They merely predict words and their combinations. Neurophysiologists believe that the solution to this problem lies in "human-AI" interfaces.

⚫️ Hardware Limitations

Creating AGI agents will require significant computational power, necessitating the development of next-generation GPUs and TPUs. Additionally, the situation is complicated by the prices and shortages of semiconductor chips.

Other challenges facing AGI include: ethics and trust issues, model architecture, and the training data scaling crisis.

More Reading:

❓ What is AGI?

📈 10 Facts from a Former OpenAI Employee's Report on the Future of AI

#AGI @hiaimediaen

✈️ Around the World with ChatGPT

Planning a trip is an exciting activity, but it takes time and consideration of numerous things. ChatGPT can easily become your travel agent, making your vacation more pleasurable and memorable. We share our favorite prompts for planning any trip.

🗺 Plan an appropriate itinerary

Create a detailed itinerary for a trip to [city/country name] for [number of days] days. Break down each day separately, describing the main attractions and interesting places worth visiting. Include information on operating hours, recommended visit duration, methods of transportation between locations (walking, public transport, or taxi), as well as tips on the best times to visit to avoid crowds. Add recommendations for lunch and dinner spots.

☀ Pack your suitcase according to the weather

Write a detailed weather forecast for [city/country name] for the period from [start date] to [end date]. Include daytime and nighttime temperatures, chances of precipitation, humidity, and wind speed. Also, add advice on clothing and accessories for a comfortable stay in this region during the specified period.

🗓 Find out about interesting events

Find a list of interesting events and activities in [city/country name] on the internet for the period from [start date] to [end date]. Include festivals, concerts, exhibitions, tours, and other cultural events. Provide dates, times, locations, and ticket information (if applicable).

💡 You can refine your request by specifying your travel budget or culinary preferences.

#manual @hiaimediaen

⭐️ Last Day to Pay for Telegram Bots Under Current Terms

Hello everyone! Starting June 12, all bots and apps on Telegram must switch to using "Stars" — Telegram's internal currency. You can buy Stars through in-app purchases on the App Store and Google Play.

📱 Why the Change?

The main reason is the developer rules from Apple and Google. Without complying with these rules, developers can't distribute apps on the App Store and Google Play.

These rules state that for selling digital goods or services within their apps, developers must use Apple's and Google's in-app payment systems, which come with a 30% commission fee.

More details here.

🔍 What Does This Mean?
1. Prices for Telegram services will increase by about 30%.
2. In some countries, there might be difficulties purchasing Stars through the App Store and Google Play.

💡 How to Prepare?
Purchase subscriptions for the bots and services you need today.

On @GPT4Telegrambot, you can get annual Premium and Combo subscriptions with a 35% discount and packages of up to 1000 generations for GPT-4, Claude 3, and Midjourney. You can subscribe in the /premium section.

Tomorrow, standard payment methods will become unavailable. We will continue monitoring developments and use all options Telegram's rules allow.

#news @hiaimediaen

AI Reads Your Thoughts and Creates Paintings

"The kind of artificial intelligence we are working with is not science fiction. It's mathematical algorithms" declare members of the French art collective Obvious, which works at the intersection of science and art. Their sensational work "Portrait of Edmond de Belamy", entirely generated by AI, sold for $432,500 at a Christie's auction in 2018. The portrait marked a new era of AI art.

This time, the artists created a painting using brain waves and AI. The Obvious mind-to-image project enables the visualization of human imagination. To accomplish this, the artists collaborated with Sorbonne University and used functional MRI (fMRI).

Experiment #1: Participants in an MRI were shown a dataset of 1,200 paintings generated with Versatile Diffusion and Midjourney. Their brain scans were then processed with algorithms that reconstructed visuals based on brain waves. The artists said that the results were "almost perfect accuracy." 🔼

Experiment #2: Artists in the MRI were asked to imagine a surreal portrait or landscape and then add emotions to it, such as sadness, love, or astonishment. Afterward, the participants verbally described the envisioned images to researchers for result verification. In 9 out of 10 cases, the AI succeeded in this task.

🎨 During the experiment, one of the Obvious trio members visualized a volcano. This is how he described his fantasies:

She cascades from the hollow of a volcano, her thoughts embrace the skies, and her fire sweeps across the earth, nourishing its vessels until weariness is found, and sometimes the anger falls silent.

By integrating human creativity, AI capabilities, and fMRI, the Obvious team proclaimed the birth of "neosurrealism." In their manifesto, they described it as "the fusion of the human mind and AI algorithms," capable of unlocking boundless creative potential.

The scientific experiment results can be found here.

Sources: Cornell University, Artnet News

The illustration in this post shows the result of the experiment. Does it resemble a volcanic landscape?

🔥 — Yes, very cool!
🙈 — Scribble-scrabble

#news @hiaimediaen

📣 Hello! Welcome to our Sunday digest, featuring the most exciting AI news from the 23rd week of 2024.

MAIN NEWS

📱 Apple new releases. Apple is preparing to present its biggest iOS18 update on Monday. One of the intriguing innovations is Eye-Tracking, which allows you to control your iPhone with your gaze.

BOT UPDATES

🎸 New Suno 3.5 model. Create radio-quality songs and music with @GPT4Telegrambot.

SAVE THIS — IT'S HELPFUL

🎞 Pika Art is an AI model for generating videos based on text descriptions and images.

STARTUP

🛳️ AI is a ship’s captain! The startup uses AI to develop safe, autonomous navigation.

TO READ

🚀 Elon Musk vs Jeff Bezos: who will reach the Moon first? The two richest men on the planet plan to land astronauts on the Moon by 2029.

📈 10 insights on the future of AI. A former OpenAI employee published a report predicting the emergence of Artificial Superintelligence and giant investments in AI.

🧀 Cheese by an astrophysicist. Astrophysicist Oliver Zahn's startup has synthesized milk protein casein using AI and now produces cheese from plant-based ingredients.

📱 On building an AI app for everything. The founder of Notion shares how he came up with the startup idea, why he showed an investor a piece of paper, and what AI features you can find on Notion.

TO LISTEN

🎙 How AI is changing the music industry. GRIMES invites fans to record tracks with her synthesized voice, DRAKE uses AI to recreate the voices of Tupac and Snoop Dogg, and Taylor Swift fans are questioning the authenticity of her latest album.

TO WATCH

🎬 In the movie "Furiosa," AI is used to alter the faces of actresses playing Furiosa at different ages for greater continuity.

TO PLAY

🤖 7 games about a future where humans live alongside AI. We've gathered the most engaging games for you.

Have a great end to your week 🙂

#AIweek | @hiaimediaen