An Irish newspaper has apologized to its readers after an opinion piece on its website was discovered to be a fake, generated in part with artificial intelligence, according to the Guardian. Notably, last week, the Irish Times published an online editorial piece with the headline ”Irish women’s addiction with artificial tan is troubling.”
Adriana Acosta-Cortez, a 29-year-old Ecuadorian health worker living in north Dublin, was identified as the author. The author examined the widespread usage of artificial tan in Ireland, claiming it was a kind of cultural appropriation and disrespectful to people of color in the piece.
According to the Independent, the essay was the second-most-read article in the paper and sparked controversy on radio and social media. Several social media users also questioned whether the writer’s photo and name belonged to a real person. Many speculated that the byline image of a blue-haired Latina woman was generated by artificial intelligence.
The article was withdrawn the same day and is now permanently gone from the website. Later, the newspaper issued a statement in which it explained the situation to its readers. Editor Ruadhan Mac Cormaic said they had been tricked by the author after falling prey to a “deliberate and coordinated deception.”
According to the Guardian report, the person used GPT-4 for creating roughly 80% of the article and the image generator Dall-E 2 to create a profile photo of a journalist. “It was a violation of trust between The Irish Times and its readers, and we sincerely apologize. The incident exposed a flaw in our pre-publication protocols. We need to strengthen them, and we will,” the publisher said.
Cohere has created an LLM University to teach about Natural Language Processing (NLP) and Large Language Models (LLM). The proposed curriculum aims to give a rock-solid foundation in NLP, equipping one with the skills needed to develop their own applications.
Whether learning semantic search, generation, classification, embeddings, or any other NLP technique, this curriculum is suitable for all. The LLM University caters to learners from all backgrounds, covering everything from the basics to the most advanced topics in large language models (LLMs), ensuring one can harness the full potential of LLMs. Plus, learners have the opportunity to work on hands-on exercises, allowing them to build and deploy their very own models.
The curriculum covers LLMs starting from the basics, all the way to building and using text representation and text generation models. Its theoretical portion is explained clearly and with analogies and examples rather than formulas, and its practical portion contains lots of useful code examples that will help you solidify your knowledge.
It is a one-size-fits-all approach, and learners can pick their own path. For this reason, it is geared for anyone excited about NLP, from beginners in ML, any developer looking to build apps with language AI as well as advanced learners who are ready to put their skills into practice.
The curriculum will include topics such as working of LLMs, applications of LLMs, use of LLMs to build and deploy apps, and building apps, including semantic search models, text generators, etc. The university will also be conducting reading groups and hosting events exclusively.
The Coca-Cola Company has forayed into the field of generative AI by using the technology to create its latest advertisement. Coca-Cola launched an AI-powered campaign, Masterpiece, that has taken the world by storm, showing some of history’s most renowned artworks while using breakthrough artificial intelligence.
This creative advertisement follows a Coca-Cola bottle as it travels through some of the world’s most iconic paintings and sculptures on its way to a thirsty student in need of inspiration. The VFX team at Electric Theatre Collective and creative firm Blitzworks collaborated to produce a magnificent commercial that flawlessly transitions between several forms of artwork using a mix of live-action images, digital effects, and AI.
The advertisement contains some of the world’s most famous artists, including Utagawa Hiroshige, J.M.W Turner, and Vincent Van Gogh, as well as modern artists such as Aket, Vikram Kushwah, Stefania Tejada, and Fatma Ramadan.
Coca-Cola’s Masterpiece campaign will include 3D billboards and digital collectibles in addition to the two-minute commercial. On its Masterpiece Gallery website, the brand has curated the art used in the advertisement, as well as talks with the featured modern artists. Wonderbuhle, a South African artist whose work appears in the advertisement, says Coca-Cola reminds him of “joy and coolness.”
Coca-Cola announced in February 2023 that it would collaborate on marketing efforts with OpenAI’s DALL-E2 model and ChatGPT. It was Bain’s first deal since the latter launched a collaboration with OpenAI earlier this year, primarily to use its tools for marketing.
Oracle is well-known for its cloud products and services. There are many organizations worldwide that use Oracle applications. Oracle HCM is one such tool that is used to manage the overall process from employee hiring to retirement. This tool can be easily accessed in on-premise and cloud environments.
Though the fact cannot be denied that there are quarterly updates released by Oracle for the addition of new features and enhancements for the users. For instance, Oracle’s 22D quarterly update has already been released and it corresponds to financial SCM and HCM modules. For the effective implementation of these updates, a robust testing strategy is required. Oracle Cloud testing tools are used for carrying out the testing.
Understanding the Oracle HCM Software
Oracle HCM (Human Capital Management) software is a holistic suite of technologies that support businesses in managing employees from hiring to recruitment.
There are various ranges of HR functions offered by HCM software, such as payroll, recruitment, and talent management. The HCM software manages the employee data in the database and it can be run and accessed either in the on-premise or cloud environment.
What Businesses Achieve Through Oracle HCM
Through Oracle HCM, business can accomplish the following things:
Workforce Planning
The planning and management of software can be done excellently through Oracle HCM software. For example, the senior manager can easily model the business strategy and needs of the workforce through this software. Oracle Cloud testing tools can be used to verify and validate the software as per the input.
Execution and Skill Management
As there are so many employees in an organization, the evaluation of their performance becomes a tedious and time-consuming task for the team and line managers. Apart from this, the feedback process and setting up of future objectives may have to suffer if done manually. This is why the automation in HCM software helps to store the feedback. Moreover, the performance evaluation can be easily done through the performance management system module of HCM.
Better Policy Implementation
There are various policies and compliance associated with the organization’s workflow. However, their implementation may suffer due to a lack of proper resources and transparency. For example, if a new employee has joined an organization then the proper analysis of policies and other workflows can be provided through Oracle HCM software. The data can be analyzed well and salary can be tracked easily. One can easily check the compensation spent, sliced in multiple ways. Furthermore, business analysts of particular organizations can use this data to build the future compensation strategy.
Development and Learning Task
The learning and development strategy can also be implemented through this software. The different types of information related to different courses or training certificates can be stored. Furthermore, it can also assist with certification to the people outside of the organization.
What Benefits Businesses Can Avail from the Automation Testing of Oracle
Oracle HCM is quite a beneficial software. As there are so many functionalities and related updates associated with it, automation testing becomes critical. Oracle Cloud testing tools are used to streamline the testing process. Below given are the benefits of automated Oracle HCM:-
Consumes Less Time
Due to automation, test scripts can be generated speedily and executed parallelly as well that too with little human assistance. Less amount of time is required for bug and error fixation. Moreover, automated testing also streamlines continuous and shift-left testing that enables the testing team to find bugs or errors in the early development phase.
Enhanced Efficiency
Automation permits testing around the clock without human intervention. The process of testing gets streamlined within the Oracle Cloud software, due to which the focus can be put on essential activities. The use of automated Oracle Cloud testing tools reduces the chances of human error and hence delivers more reliable results in comparison to manual testing.
Risk Management
Through automation testing, security vulnerabilities can be easily identified, hence making it easier to address before the arrival of the problem. The organization can quickly test its critical business application with automation testing. Moreover, the frequency of testing speed and quality can also be increased.
Compliance Maintenance
Automated tests can be run regularly, which helps ensure whether the system is operating correctly or is aligned with the given input and set standards. The automated Oracle Cloud testing tools enable the HCM software validation through the configuration of set security standards.
Improved Productivity
Through automation testing, the amount of time gets reduced to a large extent. The employees need not invest much of their time in executing the same test script. Businesses that embrace the technique of automation get a high retention rate of employees because automation minimizes the frequency of doing the same task again and again.
Final Thoughts
HCM is one of the well-known and widely used software of Oracle. It helps streamline the overall workflow of the organization, which comprises employees. Though there are some challenges also associated with it but with the right strategy, it can easily be overcome.
Opkey is a no-code test automation Oracle cloud testing tool through which testing can be carried out seamlessly. To know more, you can visit their website.
To assure moral and practical AI development, the Indian Institute of Technology Madras (IIT Madras) established the Centre for Responsible AI (CeRAI).
According to a press release, CeRAI will concentrate on interdisciplinary research in Responsible AI and seeks to become a premier research institution with immediate influence on the deployment of artificial intelligence systems in India. Today, May 15, the Centre for Responsible AI held its inaugural session on “Responsible AI for India.”
Abhishek Singh, managing director, chief executive officer, Digital India Corporation, said, “AI is significantly influencing all of our lives. Whether we are aware of it or not, we use AI-based technologies every day in some aspect of our lives.”
“The risks and challenges that still exist while we use the same technologies for solving societal problems, ensuring access to healthcare, making healthcare more affordable, making education more inclusive, and increasing agricultural productivity, are extremely important than those involved in policy making and those working at the cutting edge of developing technologies,” he added.
The main objective of CeRAI is to produce research outputs, such as patents, white papers, and research articles that are published in high-impact journals and conferences. It also aims to produce technical tools in the form of toolkits, software, and curated datasets for the Responsible AI industry.
The center has organized a panel debate on “Responsible AI for India” during a workshop and seeks to collaborate with businesses, academic institutions, and government agencies. According to the statement, CeRAI is currently actively involved in attaining its goals.
The beta release of OpenAI’s eagerly awaited ChatGPT Plus features has been revealed. Subscribers may now use ChatGPT plugins, enhancing the functionality of this strong tool. Users of ChatGPT Plus now have access to over 70 third-party plugins, which were previously only available to a small group of users during the alpha period.
“We’re bringing out web browsing and Plugins to all ChatGPT Plus members over the coming week,” OpenAI said on Twitter. “From alpha to beta, they let ChatGPT employ more than 70 third-party plugins to access the internet. With this upgrade, customers will have the exciting chance to assign ChatGPT a variety of duties, which will increase its usefulness.”
Users that sign up for ChatGPT Plus will immediately have access to the plugins, which are easily found in the “Beta Features” area of the user settings. Users who purchase this $20 per month subscription get access to the ChatGPT-4 model, which, although not perfect, is a major improvement over ChatGPT-3.
The addition of more than 70 third-party plugins gives a variety of features, such as data visualization and analysis, language learning support, and website content processing.
Users are free to install as many plugins as they like, but they are limited to using three at once. These plug-ins provide customers a wide range of capabilities and cater to a variety of businesses, including retail, real estate, stocks, travel, and cuisine. For instance, users may quickly search and gather pertinent sports news and information with the help of the “ndricks Sports” plugin.
ChatGPT becomes a flexible tool for a variety of purposes when it can access the internet and use these plugins. OpenAI keeps pushing the limits of AI technology while providing consumers with creative solutions.
Microsoft researchers created a brand-new prompt optimisation technique called Automatic Prompt Optimisation (APO). A broad and nonparametric rapid optimisation technique called APO was developed in the spirit of numerical gradient descent. It attempts to automate and enhance the rapid development procedure for LLMs.
The methodology extends previously developed automated methods, such as developing auxiliary models or differentiable representations of the prompt and implementing discrete manipulations utilizing reinforcement learning or LLM-based feedback.
APO, in contrast to earlier approaches, overcomes the discrete optimisation hurdle by incorporating gradient descent into a text-based Socratic discussion. Backpropagation is replaced with LLM editing, and distinction is replaced with LLM feedback. The system begins by obtaining natural language “gradients” that explain the defects in a particular prompt using mini-batches of training data.
These gradients serve as a guidance during editing, where the prompt is changed to follow the gradient’s inverse semantic orientation. The prompt optimization problem is then changed into a beam candidate selection problem by doing a broader beam search to broaden the search space for prompts. The effectiveness of the algorithm is improved by this method.
The Microsoft research team tested APO’s performance against three cutting-edge prompt learning baselines on a variety of NLP tasks, including the detection of jailbreaks, hate speech, bogus news, and sarcasm. On all four tasks, APO consistently beat the baselines, outperforming both the Monte Carlo (MC) and the reinforcement learning (RL) baselines by a wide margin.
Leading branchless banking and digital network PayNearby has introduced a meta commerce platform to enable easy availability of products and services in an assisted mode at the last mile, ahead of the nationwide rollout of the Open Network for Digital Commerce (ONDC).
The meta platform will soon be merged with ONDC, and PayNearby intends to use the initiative to modernize every merchant in its network. Within hours of the mall’s inauguration, the business announced that over 6000 retailers had upgraded their accounts.
Over 250 orders were placed by more than 100 shops. According to PayNearBy, the most popular categories were mobile phones and accessories, home appliances, consumables, and textiles, with many people looking for two wheelers.
It aims to empower retailers and sellers, ranging from large merchants to local sellers and small enterprises, to meet the demands of their clients both locally and across a larger national market. The platform will assist shops in broadening their scope, running a more lucrative business with no additional working capital, and utilizing technology to make all types of services available at the last mile.
For clients, it aims to deliver a seamless, cheap one-stop-shopping experience in which they can go into their nearest PayNearby store and, with the assistance of a local retailer, discover any seller, product, or service at the best price. According to PayNearBy, it will enable an optimal demand and supply match, which will likely promote the expansion of local economies.
”Retailers will no longer be constrained by their working capital, store size, or shelf space. All they would need is their PayNearby/NeoDukaan app to offer international inventory to their clients at the press of a button, supported by 50+ brands/partners,” stated Founder, MD & CEO, PayNearby, Anand Kumar Bajaj.
Walden University is actively utilizing many distinct artificial intelligence models to assist students in tutoring, completing assignments, and improving their verbal and nonverbal communication abilities.
Steve Tom, Adtalem’s Chief Customer Officer, has assisted in the deployment of three unique AI systems at Walden University: Charlotte, Linda, and Julian. The tools assist counseling students in preparing for professions by allowing them to work with “digital people” to develop communication and crisis management skills.
Charlotte is a digital assistant chatbot that can assist students in staying on top of duties and assignments in order to properly navigate a class curriculum. Students can interact with Charlotte by typing queries, which the model then processes, interprets, and responds to. According to Tom, tens of thousands of pupils are using the model around the clock, resulting in a resolution rate of more than 90%.
Linda, on the other hand, is a “digital human” who assists students in practicing sophisticated counseling conversations. Linda serves as a virtual simulation for many students pursuing careers in social work, counseling, and psychology, allowing them to test their empathetic aptitude and perfect an impactful combination of verbal and nonverbal skills.
Julian, the third AI model, has been a significantly lengthier journey, according to Tom, capitalizing on the rapid advancement of large language models (LLM) and generative AI. Julian is a virtual tutor created utilizing the AI and machine learning capabilities of Google Cloud.
On Thursday, Meta unveiled “AI Sandbox” for advertisers to test prototypes of products utilizing artificial intelligence. Marketers can build and improve their commercials with the use of these artificial intelligence (AI) tools.
According to the company’s blog post, the tools will first generate different text for ads that cater to different demographics, create alternative backdrop pictures depending on the words given, and automatically resize ad images to account for platform changes. To test the tools, Meta is collaborating with a small number of marketers, and it plans to progressively open up access starting in July.
The AI Sandbox also includes new AI-based model to assist companies in forecasting the effectiveness of advertisements using a larger data set than previously. The technique, known as Meta Lattice, reportedly improved performance for marketers in early Instagram testing.
In order to revive development in its primary digital advertising sector, Meta has been under pressure. The business announced revenue growth last month after experiencing sales reductions for the first time ever for the prior three straight quarters.
That first-quarter result points to the potential conclusion of a challenging period in which a shaky economy reduced marketer spending and Apple’s modifications to iOS privacy made smartphone advertisements less successful.
