Search

Apple is set to make a big push in artificial intelligence (AI) with its platform updates this year. The company has already deployed machine learning technology in various areas, but the focus now is on generative AI. One potential area for generative AI is Siri, Apple's virtual assistant, which has not advanced much in recent years. Apple's dedication to on-device processing may allow it to leverage users' data stores, but there may be a need for cloud-based technologies for more power. Apple's AI ambitions have the potential to benefit various areas of its platforms, such as email summarization, image editing, and code generation. The company has the opportunity to come up with innovative ideas that can have a significant impact on device interaction. [Extracted from the article]

OpenAI has announced the release of GPT-4o, a new language model that can interact with users through text, voice, and visual prompts. The model can recognize and respond to screenshots, photos, documents, and charts, as well as facial expressions and handwritten information. It has improved response times, matching human conversation speeds, and offers better performance in vision and audio understanding. While some analysts believe OpenAI is catching up to competitors, GPT-4o demonstrates impressive conversational capabilities and multilingual proficiency. The model will be rolled out iteratively, with extended access for developers and new features for paying users. [Extracted from the article]

To its credit, I was impressed that Windows Copilot answered a question about using the Windows registry to modify the Windows 11 taskbar - but it didn't actually perform those modifications, and I'm not sure it ever will. Windows Copilot, the migration of Microsoft's AI-powered Bing Chat into Windows, debuted in late June as part of the latest Windows Insider build for the Dev Channel (fave.co/46KXFPW). DEPARTMENTS Windows Copilot is one part Bing, one part Windows. [Extracted from the article]

According to a survey conducted by IDC, one-fifth of organizations plan to use a chatbot assistant other than Microsoft's Copilot within the next two years. Chatbots, which are AI-powered assistants, have become increasingly popular for automating tasks and improving productivity. These chatbots can integrate with enterprise systems like ERP and CRM to automate data-intensive tasks. While Microsoft's Copilot currently dominates the market, other vendors such as Slack, Salesforce, Oracle, SAP, and Cisco are also developing their own chatbot assistants. The survey also revealed that Amazon's CodeWhisperer and Google's Generative AI for Workspaces are popular choices among users. However, Microsoft's Copilot products, including GitHub's variant, are still widely used, particularly for software code production. [Extracted from the article]

Apple is reportedly working on a deal with OpenAI's ChatGPT to bring generative AI to iPhones and Siri. The deal may also extend to other generative AI services, such as Google Gemini. Apple's challenge will be to maintain user privacy and security while supporting third-party chatbots. The company hopes to improve Siri's contextual understanding and is exploring various strands of AI development, including on-device genAI, cloud-powered AI, and chatbot services. Apple sees the use of genAI as a way to protect its iPhone kingdom and potentially supplant iOS as the primary operating system. The company is also developing its own AI solutions and new silicon for data centers. The frequency of reports on Apple's AI plans suggests that new components will be introduced at the WWDC event. [Extracted from the article]

WORKING MAC As an editor who's been covering Apple for a long time, I'm often asked by family, friends, and acquaintances about which Mac to buy. Portability: If you plan to use your Mac on the go, a laptop like the MacBook Air or MacBook Pro may be a better choice than a desktop model like the iMac or Mac Pro. - ChatGPT, April 12, 2023 The model ChatGPT recommends - an M1 Max 16-inch MacBook Pro - is another recommendation for a Mac that's discontinued. ChatGPT can give you advice on buying a Mac - the wrong advice. [Extracted from the article]

FEATURES ChatGPT (fave.co/3wnuJwM), a powerful language model chatbot developed by OpenAI, has revolutionized the way we interact with artificial intelligence. DRAFT A PERSONALIZED WORKOUT PLAN ChatGPT can create personalized workout plans (fave.co/3zZKBqN) tailored to your fitness level, goals, available equipment, health condition, and more. Just tell ChatGPT you want to play, specify the topic or theme you'd like to focus on, and then ChatGPT will generate questions for you to answer. [Extracted from the article]

AI chatbots aren't search engines - they're crypto bros But Microsoft and Google are causing just as much harm by associating chatbots with search engines. DEPARTMENTS Over the last few months, AI chatbots have exploded in popularity off the surging success of OpenAI's revolutionary ChatGPT - which, amazingly, only burst onto the scene around December (fave.co/3wnuJwM). [Extracted from the article]

PROBLEM-SOLVING: HOUSEHOLD BUDGET ANALYSIS Winner: ChatGPT, reluctantly All three chatbots answer this question in a mediocre manner: Bing doesn't provide that much context and lacks specific dollar numbers. Here are the questions we asked all three AI chatbots (Bing, then ChatGPT, then Bard), their responses, and any notes we made about the conversation. CREATIVITY: SEINFELD MEETS ELON MUSK Winner: Bing, ChatGPT Chatbots are supposed to be able to use their AI powers to emulate literary styles. CREATIVITY: A STEAMY PIECE OF FICTION Winner: Bard, by default None of the chatbots allows you to create not safe for work (NSFW) scenes, but we wanted to see what would happen when we asked all three AI chatbots to concoct a romance novel involving "Fernando" and a potential love interest. [Extracted from the article]

WELCOME TO BING CHAT If you're familiar with ChatGPT, the Bing interface won't feel all that different. DEPARTMENTS Bing's new AI-powered chatbot (see page 7) is basically ChatGPT with ads…and one that refuses to do your homework for you. With ChatGPT, the new Bing wants to be your "AI-powered copilot for the web". [Extracted from the article]

With ChatGPT, the new Bing wants to be your "AI-powered copilot for the web" Microsoft plans to have a new Bing search and Bing browser use chat technology to answer questions posed by users. DEPARTMENTS Microsoft has unveiled the "age of AI", with new conversational ChatGPT features for Bing that the company described as the "AI-powered copilot for the web.". [Extracted from the article]

DEPARTMENTS Someone on the internet wrote that ChatGPT just made the first two years of high-school homework meaningless. OpenAI's ChatGPT explainer (fave.co/ 3WloHYp) describes how humans trained ChatGPT by simulating questions and responses, and probing existential questions get shot down quickly. [Extracted from the article]

If clicking a menu bar icon still isn't fast enough, you can assign a keyboard combination to open MacGPT. MACUSER REVIEWS The Latest Mac Products Reviewed & Rated ChatGPT (fave.co/3Mbv7aN) made its debut last December, and this AI-based text-generation tool opens up a world of possibilities not just for anyone doing research on the web, but also for anyone who needs to write about any topic. [Extracted from the article]

My struggles highlight two lessons, however: First, sites like Excel Formulator and ExcelFormulaBot don't necessarily highlight what AI models they've been trained on - and probably should, to indicate which is the most sophisticated. The site is still live, but it's evolved: Now ExcelFormulabot only offers five free formula requests per month, while anything else costs $6.99 per month (fave.co/3kkO4fr). Clueless about how to write Excel formulas?. [Extracted from the article]

OpenAI quickly resolved an issue that caused an outage for some users of its AI chatbot, ChatGPT. Users took to social media to report the problem, but the company fixed it within a few hours. While occasional outages are expected with cloud-based services like ChatGPT, OpenAI's response to this incident will be closely watched as it could impact the company's future success and the safety of AI technology. It is important for OpenAI to respond quickly and take proactive measures to minimize future outages. [Extracted from the article]

Apple has been criticized for being slow to develop its generative AI (genAI) strategy, particularly in relation to its virtual assistant Siri. However, recent leaks suggest that Apple is planning to revamp Siri and license OpenAI's ChatGPT to improve its capabilities. Apple is also expected to announce AI-related developments at its Worldwide Developers Conference (WWDC) in June, including licensing complementary AI models and developing its own large language models (LLMs) for chatbots. Additionally, Apple researchers have developed a new AI system called ReALM, which improves natural interactions with voice assistants. While Apple may have fallen behind in the AI race, it is taking steps to catch up and improve Siri. [Extracted from the article]

The article discusses the potential drawbacks of generative AI (genAI) tools such as Google AI Overviews and OpenAI GPT-4o. The author highlights that these tools, while capable of producing convincing results, are still prone to errors and hallucinations. The accuracy of genAI tools depends on the quality of their sources, which can vary widely. The author also raises concerns about the impact of AI-powered summaries on publishers, as it may discourage users from accessing the full content and negatively affect the industry. Additionally, the article mentions the potential decline in the quality of AI models as reliable human-created stories become scarce. The author concludes by emphasizing the need to address these issues to prevent a future filled with unreliable AI-generated content. [Extracted from the article]

Ilya Sutskever, co-founder and chief scientist of OpenAI, has announced his departure from the company. He will be replaced by Jakub Pachocki, the former director of research. Sutskever's departure comes after a period of conflict within OpenAI, during which he and other board members pushed out CEO Sam Altman, only to have Altman return shortly after. While Altman expressed sadness at Sutskever's departure, there are concerns about the future of OpenAI and its customers without Sutskever's vision and leadership. However, Sutskever himself is optimistic about his next venture, which he plans to share details about in the future. [Extracted from the article]

Anthropic, a competitor of OpenAI, has launched its AI chatbot named Claude in Europe. Claude supports multiple European languages and is available as a web-based version, an iOS app, and a Team plan for businesses. The expansion to Europe comes amidst the region's strict regulatory environment for tech companies. Anthropic's entry into the European market is expected to disrupt the existing balance and foster a competitive environment that could benefit European businesses. [Extracted from the article]

A recent study conducted by career development and outplacement firm Intoo, in partnership with research firm Workplace Intelligence, found that many Gen Z employees trust AI chatbots like ChatGPT for career advice over their managers. Around 47% of Gen Z workers surveyed said they received better advice from a chatbot than from their manager. The survey also revealed that toxic work environments and a lack of career advancement are contributing to employee dissatisfaction. Additionally, the study highlighted that many workers are actively seeking employment elsewhere due to poor treatment and a lack of support for career development. [Extracted from the article]

Google has expanded access to its generative AI (genAI) features within its Workspace office app suite, offering a lower-priced payment tier aimed at small and mid-sized businesses. The new pricing tier, called Gemini Business, costs $20 per user per month and provides most of the same capabilities as the higher-priced Gemini Enterprise tier. Both tiers also include access to Google's new standalone chatbot AI assistant, Gemini. This move puts Google's genAI offerings more in line with competitors like Microsoft and OpenAI. However, convincing small and mid-sized businesses to invest in genAI tools may be challenging due to the added cost and potential difficulty of use. [Extracted from the article]

OpenAI has introduced a new memory feature in ChatGPT, their chatbot, which allows it to remember information about users and past conversations. This feature aims to personalize conversations and make the chatbot more user-friendly and intelligent. It is expected to increase productivity and enhance customer experience for organizations. Users have control over what the chatbot remembers, and they can also choose to use a "temporary chat" feature without memory. While the memory feature raises data privacy and security concerns, some believe that these concerns are more relevant to consumers rather than enterprises. [Extracted from the article]

The article discusses the author's experience with Google's new Gemini Android assistant and expresses skepticism about its effectiveness as a replacement for Google Assistant. The author argues that Gemini, as a generative AI chatbot, lacks a clear purpose and does not fulfill the basic tasks and functions that users expect from a phone assistant. The article highlights the slower and less reliable performance of Gemini compared to Google Assistant, as well as its inability to handle certain tasks. The author suggests that Google should focus on improving Google Assistant rather than replacing it with Gemini. [Extracted from the article]

Large language models (LLMs) are the foundation of chatbots like OpenAI's ChatGPT and Google's Bard. LLMs are trained on massive amounts of data, including articles, books, and internet resources, to generate human-like responses to natural language queries. They work by predicting the next word based on what they have already seen. LLMs can be biased if the data they ingest is biased, and prompt engineering is a process that allows users to train LLMs for specific purposes. Smaller LLMs are being developed to reduce costs and improve efficiency. However, there are concerns about privacy, security, and ethical use of data collected by LLMs. Efforts are being made by governments and organizations to regulate and govern AI tools. [Extracted from the article]

Italian watchdog Garante has accused OpenAI's generative AI platform, ChatGPT, of breaching data privacy norms set by the EU. This is not the first time that Garante has brought charges against ChatGPT, as last year the watchdog temporarily banned the platform for flouting data privacy norms. OpenAI has been given 30 days to respond to the charges. These accusations come at a time when OpenAI is already facing multiple lawsuits regarding the training of its models without explicit permission from content creators. [Extracted from the article]

OpenAI has responded to The New York Times' copyright lawsuit, claiming that the examples of copyright violations provided by the newspaper were cherry-picked and manipulated. OpenAI argues that their models do not typically behave in the way the NYT suggests and that the generated texts are not a substitute for the newspaper's content. OpenAI acknowledges the issue of "regurgitation" or memorization in their model and is working to address it. Some experts, including AI scientist Andrew Ng, sympathize with OpenAI's position, arguing that training AI models using publicly available internet materials should be considered fair use. [Extracted from the article]

OpenAI, a leading AI research organization, has argued that the development of advanced generative AI tools is not possible without the use of copyrighted content for training. In a report to the UK's House of Lords Communications and Digital Select Committee, OpenAI stated that training large language models like GPT-4 would be impossible without copyrighted materials. However, this assertion has sparked a legal debate over intellectual property rights and the use of copyrighted content without credit or compensation. OpenAI is currently facing multiple lawsuits, including one from The New York Times, alleging illegal use of copyrighted content in the creation of their tools. [Extracted from the article]

Athenahealth, a healthcare technology provider, has been using AI for years to improve efficiency in processing healthcare requests and records. Recently, they adopted generative AI (genAI) through OpenAI's ChatGPT platform, which has allowed them to introduce new capabilities across their product line. These capabilities include intelligent summarization of patient healthcare documents and identifying missing or incorrect information before submitting a prior authorization for care. Athenahealth has invested in educating their employees on AI and has created an AI team to explore new ways of utilizing the technology. They have also taken measures to ensure the security and privacy of their sensitive healthcare data when working with AI models. [Extracted from the article]

According to a report by Cloudflare, global internet traffic increased by 25% in 2023, indicating a growing reliance on online services. Google remained the most popular website, followed by Facebook, Apple, and TikTok. OpenAI emerged as the leader in generative AI services, with other notable players including Character AI, Quillbot, and Hugging Face. Financial organizations were the primary targets for cyberattacks, with deceptive links and extortion attempts in emails being the most common threats. Internet outages also increased, often due to government-imposed shutdowns. Smartphones were the preferred device for accessing the internet, with Apple devices being popular among users. Generative AI became a significant trend, with OpenAI's ChatGPT launch gaining attention. Cloudflare used its complimentary service, Cloudflare Radar, to gather data for the report. Elon Musk's Twitter, now known as X, experienced a decline in popularity throughout the year. Internet outages increased, with over 180 incidents reported globally. Security concerns included vulnerabilities like Log4j and HTTP/2 Rapid Reset, as well as ransomware, malware, and phishing attacks. [Extracted from the article]

A new study by Kaspersky Research suggests that generative artificial intelligence (genAI) could help address skills shortages in the job market. The study found that 40% of surveyed C-level executives plan to use genAI tools like ChatGPT to automate tasks and fill critical skills gaps. However, 91% of respondents admitted to not fully understanding how genAI works. While genAI offers benefits, there are concerns about high costs, uncertain return on investment, and potential exposure of sensitive data. Forrester Research predicts that genAI will replace 90,000 jobs in 2023 and 2.4 million jobs by 2030, but job losses will be lower than expected. [Extracted from the article]

Generative AI platforms, such as ChatGPT, are highly inaccurate and becoming less transparent when connected to corporate databases, according to two studies. The Stanford University study found that as these platforms ingest more data and grow in size, it becomes harder to track the origin of the data they use. This lack of transparency makes it difficult for businesses to build applications using commercial genAI models and for academics to rely on them for research. Additionally, the overall accuracy of these models has been questioned, with one study showing that they return accurate responses to basic business queries only 22% of the time. The inaccuracy of these models erodes trust and can have significant consequences for businesses. To address these issues, companies need to invest in strong data foundations, such as Knowledge Graphs, and ensure oversight and governance to protect sensitive information. [Extracted from the article]

A group of nonfiction authors has filed a class-action lawsuit against OpenAI and Microsoft, accusing them of copyright infringement. The authors claim that their copyrighted materials were used without their consent to train the AI chatbot ChatGPT. The lawsuit alleges that OpenAI and Microsoft have ignored copyright laws for their own financial gain. The plaintiffs are seeking damages, restitution, and a permanent halt to the infringement. This lawsuit is part of a larger trend of legal challenges against the use of copyrighted materials in AI platforms. [Extracted from the article]

The firing of Sam Altman, co-founder of OpenAI, has caused turmoil within the company and raised concerns about the future of generative AI (genAI) technology. OpenAI employees have threatened to quit and work for Microsoft unless the entire board of directors resigns and reappoints Altman as CEO. However, industry experts believe that even if OpenAI were to collapse, it would not significantly impact AI development, as many other well-funded companies and startups are pursuing generative AI. It is advised for companies to build their AI strategies responsibly and not overly rely on a single provider. The conflict within OpenAI highlights the need for global regulation for safe and secure AI. [Extracted from the article]

New York Life Insurance Co. has been using predictive models and artificial intelligence (AI) to improve efficiency and productivity since 2015. The company has deployed AI in various areas, including actuarial assumptions, agent recruiting, and marketing campaigns. In November 2022, the company adopted ChatGPT, a language model-based chatbot that has significantly enhanced customer service and training for employees. New York Life is focused on addressing challenges related to AI, such as ensuring the technology is used in areas that make a difference, managing talent and change, and implementing ethical AI practices. The company believes that AI will primarily augment human workers rather than replace them, and it is actively preparing its employees for the changing nature of their roles. [Extracted from the article]

The chatbot, which has been christened Grok AI, wasdeveloped by xAI - a generative AI venture launched byMusk in July to make an AI model understand the truenature of the universe in order to make it safer. AI safety continues to be a major concern Musk's decision to launch xAI and subsequently Grok AIcould be seen by many to be contradictory to hisprevious remarks. Elon Musk has revealed a new generative AI model-basedchatbot that will take on several large language modelsincluding OpenAI's ChatGPT, Google's PaLM 2, andAnthropic's Claude 2. [Extracted from the article]

News Generative AI chatbot ChatGPT can keep up with currentevents thanks to a new update, and is also getting directaccess to the DALL-E 3 image-generation system The latest version of ChatGPT, OpenAI's groundbreakinggenerative AI chatbot, now has access to real-time informationfrom the internet at large, curing what had been seen as oneof its key flaws. Thanks to an integration with Microsoft's Bing search engine- the company is a major investor in ChatGPT maker OpenAI- the AI chatbot is now able to look up the latest availableinformation on any topic. [Extracted from the article]

Google today announced it is embedding its conversational,Bard chatbot into many of its most popular apps, allowing users to get human-like responses to questions, summarize Gmail messages, and search Google Docs and Youtube. Microsoft has added ChatGPT functionality to its suite of Microsoft 365 business productivity and collaboration apps through its own Copilot chatbot assistant. [Extracted from the article]

These include the AI models that were recently unveiled by China's largest search engine, Baidu, and AI software company SenseTime. Chinese cloud and social media conglomerate Tencent has announced that its Hunyuan large language model (LLM) will be available for enterprise use, as companies in the country stake their claim as technology industry and generative AI leaders. [Extracted from the article]

In this Q&A, Computerworldposed a series of questions via email to Trachet about ChatGPT Enterprise and how it could affect the adoption of genAI tools at both startup firms and more established enterprises. Feature ChatGPT Enterprise will likely spur greater adoption of generative AI technology by addressing two of the main concerns businesses have: security and privacy. [Extracted from the article]

In addition to controlling the use of ChatGPT and other standalone tools, companies now have to grapple with generative AI being built into the productivity apps their employees use every day Companies have already started on this journey, elevatingdata literacy among their employees as part of their push tobecoming a data-driven enterprise, said Omdia's Shimmin." DLP (data loss prevention) tools can help companies keepsensitive data from leaving the company. Often, a company's security team might not even know that aparticular application used by the company has now addedgenerative AI tools, he said. [Extracted from the article]

OpenAI customers will be able to securely extend ChatGPT'sknowledge with a company's data by connecting it to theapplications they already use, but that feature is still beingdeveloped, according to a spokesperson. News Analysis OpenAI said the newest version of its ChatGPT servicecomes with enterprise-grade security and privacy features OpenAI today unveiled ChatGPT Enterprise, a version of itsgenerative AI service that allows companies to decide how themodel is trained and how long it can store corporate data usedfor that purpose. OpenAI launches "enterprise-grade" ChatGPT. [Extracted from the article]

During the past few months, Microsoft has been integrating Copilot widely across its various products, with versions of the AI assistant tailored to Dynamics, Microsoft 365, Power Platform, Security products, and others currently under development. But the effects of automation will be felt more widely, says Microsoft, which is bringing its Copilot AI assistant to applications used by those in frontline roles, whether that's retail workers or technicians out in the field. Microsoft 365 Copilot. [Extracted from the article]

The article discusses the popularity of OpenAI's ChatGPT, the first widely-used generative AI tool. However, the author suggests that it may not be the best option and encourages readers to explore other powerful alternatives. The article explains that there are thousands of comparable models available, with Stanford scientists creating a resource called Constellation to visualize and understand them. The author highlights Microsoft and Meta's LLaMA 2 as a new LLM offering and mentions that Apple has reportedly developed its own LLM called "AppleGPT." The article also emphasizes the importance of using multiple sources and tools to verify information obtained from LLM-based chatbots, as they can be prone to errors and hallucinations. The author recommends experimenting with various AI tools and using AI as a writing partner to enhance writing and thinking skills. [Extracted from the article]

News OpenAI is reportedly under additional legal scrutiny, asthe US Federal Trade Commission asks the company to givedetailed explanations of its business practices The US Federal Trade Commission is reportedly investigatingChatGPT maker OpenAI for potential violations of consumerprotection regulations involving data privacy and falsestatements about third parties. [Prepare for generative AI with experimentation and clearguidelines] While the FTC declined to comnent, OPenAI CEO Sam Altman issued aresponse to the reports on Twitter. [Extracted from the article]

News Five months after Google announced it was launching its owngenerative AI chatbot, the company has addressed the data privacyissues that had previously made it inaccessible to EU residents. Like other generative AI chatbots, Bard is powered by a largelanguage model (LLM), in this case a lightweight variant ofLaMDA, Google's main natural-language processing model. [Extracted from the article]

News Five months after Google announced it was launching its owngenerative AI chatbot, the company has addressed the dataprivacy issues that had previously made it inaccessible to EUresidents. Like other generative AI chatbots, Bard is powered by a largelanguage model (LLM), in this case a lightweight variant ofLaMDA, Google's main natural-language processing model. [Extracted from the article]

OpenAI is opening a new alignment research division, focused on developing training techniques to stop superintelligent AI -- artificial intelligence that could outthink humans and become misaligned with humans ethics -- from causing serious harm. OpenAI acknowledges the need to mitigate potential AI harm This isn't the first time OpenAI has publicly acknowledged the need to mitigate the risks posed by unregulated AI. [Extracted from the article]

"Because the OpenAI Language Models cannot functionwithout the expressive information extracted fromPlaintiffs' works (and others) and retained inside them,the OpenAI Language Models are themselves infringingderivative works", according to the complaint in thatcase, also filed in the Northern District of California. News A new lawsuit against OpenAI could decide whetherthe company's use of training data scraped from thepublic internet may continue OpenAI, the Microsoft-backed company that developed theChatGPT generative AI chatbot, is the defendant in afederal class action lawsuit filed this week inCalifornia, where it is accused of misappropriatingpersonal information for training purposes. [Extracted from the article]

News AI-transcription service Otter.ai has launched anew AI Chat feature with the ability to interactwith multiple meeting participants in real time. AI-transcription company Otter.ai has announced the launch of OtterAI Chat, a new AI-powered meeting tool that can provide real-time,question-and-answer support and generate meeting-specific content. "This is why we have designed a new chat model that supportsmultispeaker AI chat, making AI a participant and getting it involvedin the conversation", he said. [Extracted from the article]