AI Archives - Page 51 of 51 - Jet Developers Blog

GPT-4 to have 100 Trillion Machine Learning Parameters

Microsoft plans to release GPT-4 as early as next week, with the ability to create AI-generated videos from simple text prompts.

Andreas Braun, Chief Technology Officer at Microsoft Germany, recently confirmed that GPT-4 will be unveiled next week at an event called AI in Focus – Digital Kickoff, reports Windows Central.

“We will introduce GPT-4 next week, where we have multimodal models that will offer completely different possibilities – for example, videos,” Braun was quoted as saying.
The report said that it is the next iteration of OpenAI’s Large Language Model (LLM), and it should be significantly more powerful than GPT-3.5, which powers the current version of ChatGPT.

GPT-4 which is 500 Times More powerful than the current #ChatGPT will be Released next week. The current version of ChatGPT is built on GPT 3.5 with 175 Billion Machine Learning Parameters. But GPT-4 has 100 TRILLION ML PARAMETERS.

GPT-4 will be able to process multiple types of data including Videos, Images, Sounds, Numbers etc.

ChatGPT and other GPT-3.5-powered technologies are currently limited to text-based responses.

However, Braun’s comments imply that this may change with the release of GPT-4.
The multimodal models of the LLM could pave the way for video production and other types of content, according to the report.

Meanwhile, the AI-powered Bing search engine has surpassed 100 million daily active users, as ChatGPT’s integration into Bing has helped the company grow its usage within a month like never before.

Its rival Google Search engine has more than 1 billion daily active users.

The National Security Commission on Artificial Intelligence (NSCAI) Releases Its First Quarter Recommendations

The National Security Commission on Artificial Intelligence (NSCAI) released its first quarter recommendations to maintain U.S. leadership in the development and adoption of artificial intelligence (AI). Established in the 2019 National Defense Authorization Act (NDAA), NSCAI is an independent Commission; with members representing industry, academia, and civil society organizations; that seeks to develop policy recommendations that advance the development of AI and ensure that it strengthens the American workforce, industry, innovation, values, and national security. HFES advocated for NSCAI’s creation in the FY 2019 NDAA, and has engaged with Congress and the federal agencies on a number of occasions regarding important HF/E issues that must be addressed in AI development.

These recommendations build on a previous interim report released in November 2019, where NSCAI assessed the national security landscape and global competition around AI. The Commission asserted that the future of U.S. national security and the economy depends on U.S. success in winning the AI race. The interim report described the Commission’s seven guiding principles when evaluating future recommendations and policies:

“1. Global leadership in AI technology is a national security priority.

Adopting AI for defense and security purposes is an urgent national imperative.
Private sector leaders and government officials must build a shared sense of responsibility for the welfare and security of the American people.
People are still essential. Talent remains the most important driver of progress in all facets of AI.
The power of free inquiry must be preserved.
Ethic and strategic necessity are compatible with one another.
The American way of AI must reflect American values—includes having the rule of law at its core.”

NSCAI released its first quarter recommendations in April, as Congress is examining the President’s FY 2021 budget request and crafting legislation such as the annual appropriations bills and the FY 2021 NDAA. NSCAI’s first quarter recommendations include:

Increase AI R&D Investments: The commission advocates for the federal government to double funding for AI R&D in fiscal year (FY) 2021, from $1 billion to $2 billion, in key areas such as human-AI interactions to expand AI research efforts at national laboratories and research centers, and bolster future AI applications. The commission also recommends launching a pilot program to establish a National AI Research Resource (NAIRR), in order to expand public access to large, curated data sets.
Accelerate AI Application in the Department of Defense: NSCAI recommends that DOD leadership establish mechanisms, such as a Steering Committee on Emerging Technology, to assess novel emerging technology threats and ensure that DOD is able to overcome barriers to the strategic adoption of AI and other emerging technologies. The Commission also recommends that the recently established Joint Artificial Intelligence Center (JAIC) report directly to the Secretary.
Strengthen the AI Workforce: This includes recommendations to revolutionize workforce hiring practices by targeting and identifying internal talent, establishing AI-literacy courses for human resources (HR) professionals, conducting portfolio-based rather than resume-based hiring, expanding the Cyber Excepted Service (CES), and mandating AI training, among other recommendations.
Promote U.S. Leadership in AI Hardware & 5G: NSCAI advocates for the implementation of portfolio-based approaches to advance U.S. leadership in AI and lay the groundwork for long-term AI resource and R&D investment in areas that enable AI technologies, such as microelectronics programs and fifth-generation (5G) wireless technologies and networks. This includes expanding AI-enabling microelectronics programs, developing a national microelectronics strategy, and enacting policies and programs to advance critical technical areas of 5G such as spectrum sharing.
Improve AI Cooperation Among Key Allies and Partners: NSCAI proposes that the federal government appoint a national security point of contact to deepen AI collaborations with like-minded allies and partners in order to strengthen AI-enabled warfighting, wargaming, and intelligence efforts.
Advance Ethical and Responsible AI: Acknowledging DOD’s adoption of the Defense Innovation Board’s recommended AI principles, the Commission calls for additional policies to promote the ethical and responsible use of AI by integrating ethical and responsible AI training (including understanding of AI Bias) in courses and establishing a body of experts to brief the federal government on emerging ethical issues and trends pertaining to AI.
An additional classified section with threat analysis and recommended actions was not included in the public report.

NSCAI’s report contains language for a number of legislative proposals capturing its first quarter recommendations for the NDAA, as well as recommendations for funding priorities in the FY 2021 appropriations bills. The Commission’s establishment and its work have been strongly supported by key members on the House Armed Services Committee (HASC), such as Intelligence and Emerging Threats and Capabilities Subcommittee Chairman Jim Langevin (D-RI) and Ranking Member Elise Stefanik (R-NY). It is expected that Congress will carefully consider these proposals as it prepares to work on the FY 2021 NDAA.

NSCAI will continue to release reports with recommendations as part of a framework to further AI research and development, applications, and stewardship in a dynamic technology and national security environment. The Commission notes that while the rapidly evolving threats and technology require it to remain flexible, its guiding principles that frame the recommendations have remained constant. For example, NSCAI notes that efforts to respond to the COVID-19 pandemic, including modeling and vaccine development, demonstrate the potential that AI can offer in every aspect of U.S. economic prosperity and national security.

Emirate Airlines to Introduce World’s First Robotic Check-in at Airports

A robotic check in system will be introduced by Emirates at airports in the country this year.

In what is claimed to be a world first, the system- called Sara- will speak at least six languages and help in everything from check in to hotel bookings, according to an Emirates spokesperson.

“The needs of travelers keep growing,” said Ade Redha. “However the airport is the same. We aim to serve our customers better with technology.”

Developed by Emirates in collaboration with their partners, the robotic check in system was completely locally produced. According to Redha, over 200 of these systems will be functional in airports across the city over the next few years.

AI Cancer Screening Lab put into Operation – Islamabad

ISLAMABAD, Mar. 9 (China Economic Net) – China-Pakistan joint AI Cancer Screening Lab in Islamabad was put into operation last week. The lab, located in the Akbar Niazi Teaching Hospital, will provide free cervical cancer screening to 10,000 Pakistani women in its first phase of operation.

“It can also be used in the early diagnosis of other high incidence clinical tumors, such as breast cancer, gastric cancer, oral cancer, etc.,” said the project lead from Landing Med, a Chinese medical technology company that provided three cervical cancer screening devices along with 5,000 sets of supporting consumable items to Pakistan last December for the lab construction.

Different from traditional clinical procedure where patients have to visit the hospital several times for specimen collection, report analysis, and treatment, the AI-powered lab streamlined the operation by processing the specimen, scanning the slides and uploading them to the 5G cloud platform. Once data upload is complete, a medical team in China can make diagnosis remotely and a report can be generated in about 5 minutes.

Up to now, the service has been available to surrounding residents after trial diagnosis for the hospital staff proved efficient and reliable.

Cervical cancer has become the third most common cancer in Pakistan after head and neck and breast cancers, and around 64 percent of Pakistani women who have this cancer lose their lives as they only discover the disease when it becomes almost incurable in the third or the fourth stage of cancer.

According to a study by the World Health Organization, cervical cancer incidence in Pakistan reached 4.7 per 100,000 women in 2020. However, from 2015 to 2019, fewer than 1 in 10 Pakistani women were screened for cervical cancer. Thus AI-powered, efficient and effective screening services can greatly enhance screening coverage.

China-Pakistan AI Cervical Cancer Screening Program started in 2019 on the ninth meeting of the China-Pakistan Economic Corridor Joint Cooperation Committee. Pakistan’s Ministry of National Health Services Regulations and Coordination is considering equipping more hospitals in Pakistan with such AI devices, CEN understands.

What is artificial intelligence and what is it not?

Artificial intelligence (AI) is set to transform many aspects of day-to-day life.
There are, however, many misconceptions about AI and its potential uses.
“The exaggerations about AI’s potential largely stem from misunderstandings about what AI can actually do,” said Kay Firth-Butterfield, the Head of Artificial Intelligence and Machine Learning at the World Economic Forum.

Broadly speaking, artificial intelligence (AI) is a field of study and type of technology characterized by the development and use of machines that are capable of performing tasks that usually would have required human intelligence.

AI has already transformed many industries and aspects of society, ranging from the introduction of customer service chatbots to enhanced GPS and mapping applications. However, there are several misconceptions about AI and its potential uses.

In the following Q&A, Kay Firth-Butterfield, the Head of Artificial Intelligence and Machine Learning at the World Economic Forum, details the different types of AI, important developments and applications in the field of machine learning and—perhaps most importantly—discusses common misunderstandings about AI.

What are the different types of AI?

“AI consists of several different machine learning models. These include, but are not limited to, reinforcement learning, supervised and unsupervised learning, computer vision, natural language processing and deep learning.

“All of the machine learning models develop and advance statistical predictions, but differ in their use and comprehension of data. ChatGPT, for example, is an AI-powered chatbot that is able to predict the most likely next word in a sentence. With numerous and relatively accurate predictions, ChatGPT is able to create coherent paragraphs.”

What do most people misunderstand about Artificial Intelligence (AI)?

“AI is not intelligence—it is prediction. With large language models, we’ve seen an increase in the machine’s ability to accurately predict and execute a desired outcome. But it would be a mistake to equate this to human intelligence.

“This is clear when examining machine learning systems that, for the most part, can still only do one task very well at a time. This is not common sense and is not equivalent to human levels of thinking that can facilitate multi-tasking with ease. Humans can take information from one source and use it in many different ways. In other words, our intelligence is transferable—the ‘intelligence’ of machines is not.”

Student Builds an AI Model to Translate Sign Language into English in Real-Time

Artificial Intelligence (AI) has been used to develop various kinds of translation models to improve communication amongst users and break language barriers across regions. Companies like Google and Facebook use AI to develop advanced translation models for their services. Now, a third-year engineering student from India has created an AI model that can detect American Sign Language (ASL) and translate them into English in real-time.

Indian Student Develops AI-based ASL Detector

Priyanjali Gupta, a student at the Vellore Institute of Technology (VIT), shared a video on her LinkedIn profile, showcasing a demo of the AI-based ASL Detector in action. Although the AI model can detect and translate sign languages into English in real-time, it supports only a few words and phrases at the moment. These include Hello, Please, Thanks, I Love You, Yes, and No.

Gupta created the model by leveraging Tensorflow object detection API and using transfer learning through a pre-trained model called ssd_mobilenet. That means she was able to repurpose existing codes to fit her ASL Detector model. Moreover, it is worth mentioning that the AI model does not actually translate ASL to English. Instead, it identifies an object, in this case, the signs, and then determines how similar it is based on pre-programmed objects in its database.

In an interview with Interesting Engineering, Gupta noted that her biggest inspiration for creating such an AI model is her mother nagging her “to do something” after joining her engineering course in VIT. “She taunted me. But it made me contemplate what I could do with my knowledge and skillset. One fine day, amid conversations with Alexa, the idea of inclusive technology struck me. That triggered a set of plans,” she told the publication.

Gupta also credited YouTuber and data scientist Nicholas Renotte’s video from 2020, which details the development of an AI-based ASL Detector, in her statement.

Although Gupta’s post on LinkedIn garnered numerous positive responses and appreciation from the community, an AI-vision engineer pointed out that the transfer learning method used in her model is “trained by other experts” and it is the “easiest thing to do in AI.” Gupta acknowledged the statement and wrote that building “a deep learning model solely for sign detection is a really hard problem but not impossible.”

“Currently I’m just an amateur student but I am learning and I believe sooner or later our open-source community, which is much more experienced and learned than me, will find a solution and maybe we can have deep learning models solely for sign languages,” she further added.

You can check out Priyanjali’s GitHub page to know more about the AI model and access the relevant resources of the project. Also, let us know your thoughts about Gupta’s ASL Detector in the comments below.

ChatGPT

ChatGPT is an AI Chatbot developed by Open AI. The chatbot has a language-based model that the developer fine-tunes for human interaction in a conversational manner.

Effectively it’s a simulated chatbot primarily designed for customer service; people use it for various other purposes too though. These range from writing essays to drafting business plans, to generating code. But what is it and what can it really do?

What is ChatGPT?

It is an AI chatbot auto-generative system created by Open AI for online customer care. It is a pre-trained generative chat, which makes use of (NLP) Natural Language Processing. The source of its data is textbooks, websites, and various articles, which it uses to model its own language for responding to human interaction.

This chatbot system provides information and responses to inquiries through AI. The popular version of ChatGPT is the GPT-3 model.

What is ChatGPT used for?

The main feature of Chat GPT is generating responses like those humans would provide, in a text box. Therefore, it is suitable for chatbots, AI system conversations, and virtual assistants.

However, it can also give natural answers to questions in a conversational tone and can generate stories poems and more. Moreover, it can:

Write code
Write an article or blog post
Translate
Debug
Write a story/poem
Recommend chords and lyrics

To make the AI carry out one of these demands, all you need to do is type the command into the chatbot.

What is Chat GPT Pro?

Chat GPT Pro, is a rumored professional subscription plan for OpenAI’s Chat GPT. The service was hinted to cost around $42 per month. Surprisingly, the professional plan has not yet been introduced since the reports were first spread at the start of the year.

However, OpenAI has recently released a subscription service, but under a different name – Chat GPT Plus. This plan offers users exclusive benefits including priority access and faster response times. Costing only $20 a month, this option is significantly cheaper than Pro.

What is Chat GPT trained on?

It relies on NLP (Natural Language Processing). It’s an excellent tool for researchers and developers working on various NLP projects, and it has many specific tasks, domains, and applications available to work within.

It is well-trained on biased and unbiased data, in the form of text from books, articles, and websites. Chat GPT can reproduce data outputs and reliability – crucial for many sensitive apps and other valuable Al systems. However, it is still prone to error, and biases and depends on its training data – provided in 2021.

As humans we are, more and more, interacting with Al-powered machines, and Chat GPT is a revolution in the field of Al. It is a robust model and particularly advanced thanks to its deep-learning capabilities and NLP. Ultimately, it can generate human-like answers and is easily understandable to users. Thought that doesn’t always make it right.

Can I use Chat GPT on my phone?

Yes, you can use Chat GPT on your phone. There’s nothing preventing you from doing so, as the mobile web version of the app will allow you to carry out the same actions as on a desktop browser.

Of course, you actually need a phone number for Chat GPT if you want to log in. And the ability to use it on a smartphone makes mobile use very easy – as long as you can connect.

Is there Chat GPT for Android?

There is no official Chat GPT app from Open AI, and there is no information as to whether there will be. There are purported ‘Chat GPT apps’ available in the Play Store, but these are not from the company.

Importantly, Chat GPT makes use of the GPT-3 model which is only accessible via the OpenAI Chat GPT page. But although there is no Android (or iPhone) app from OpenAI, you can access Chat GPT on your mobile device by going to the same URL.

What is Chat GPT good for?

Chat GPT is pretty good at generating text which mimics human speech. This is useful if you need a post for a website or social media page, but don’t have the time to write it out yourself. It can also produce code – again, useful if you don’t have the time to write it out yourself.