Andrej Karpathy, the genius who worked at Tesla, and now OpenAI, says that he is “building a kind of JARVIS at OpenAI.” Undoubtedly, the guy who built Baby Llama and can easily code GPT-5 over the weekend, is probably ready for the task of building autonomous agents, if not AGI, and OpenAI has possibly cracked it.
Greg Brockman, co-founder of OpenAI, recently shared the capabilities of the GPT-4 Voice where the assistant was able to do business negotiations without much supervision. Ironically, it identifies itself as Jarvis. Ammaar Reshi posted a screen recording of talking to the chatbot, where it calls itself Jarvis, and is ready to assist him with everything.
ChatGPT with Voice now refers to itself as JARVIS and knows my name thanks to custom instructions! One step closer to the ultimate assistant pic.twitter.com/NwZW7YZiH5
— Ammaar Reshi (@ammaar) October 9, 2023
There is no doubt that OpenAI has been focused on building AGI. But interestingly, the company took another step towards this goal and changed their “core values” and added the focus on AGI, something that wasn’t explicitly mentioned before on the page. Earlier, it was audacious, thoughtful, impact driven, collaborative, and so on. But now, the first is AGI, followed by intense and scrappy, scale, make something people love, and team spirit.
People have been tripping over the fact that the company changed its “core values”, but are unaware that Sam Altman has explicitly told multiple times, including other co-founders, that OpenAI is all about AGI.
OpenAI Likely to Announce “Jarvis” at DevDay
There has been a buzz going around that at the OpenAI DevDay conference, the company might just announce their first completely autonomous agent. People call it a “swarm of bees clicking around on the internet,” fundamentally changing the internet. This definitely points to the recent Global Illumination acquisition by OpenAI, where it is training AI agents in a gamified simulation.
Though the conference is focused on developers and would possibly include a lot of developer focused announcements, the release of Jarvis officially looks like in the works.
When OpenAI releases Autonomous Agents, it will be like an intelligent swarm of bees clicking around on the internet. Sending emails, negotiating, making products, purchases, fulfilling orders, etc. It will fundamentally change the internet, and this cannot be overstated.
— AI Breakfast (@AiBreakfast) October 13, 2023
Taking this forward to autonomous agents rumours, all of the recent developments indicate that OpenAI might just announce something like a Jarvis at its DevDay conference. Altman has already said in an interview that the definition of AGI for him is that something that could serve as the “equivalent of a median human that you could hire as a co-worker.”
Autonomous agents are the next step for LLM based chatbots. Almost all the companies and researchers realise this. Microsoft recently came up with AutoGEN, a framework that enables building LLM applications using multiple agents that would be able to talk to each other.
But as the GitHub repository for AutoGEN says, “AutoGen agents are customisable, conversable, and seamlessly allow human participation,” which means there’s still a need for human input for these models, which questions the autonomous behaviour of the model.
Similarly, Google DeepMind recently also published a paper – “How FaR Are Large Language Models From Agents with Theory-of-Mind?” Even Meta’s Shepherd: A Critic for Language Model Generation, talks about the same autonomous AI agents augmenting and doing tasks all by themselves. Other such papers include SELF: Language-Driven Self-Evolution for Large Language Model and SelfEvolve: A Code Evolution Framework via Large Language Models.
But none of these researches have actually been realised in reality yet though. When it comes to OpenAI, they might have just cracked AGI with Jarvis.
All roads lead to AGI
When it comes to displacement of jobs, Sam Altman had been ready for this all this while. That is why he invested in WorldCoin and wants to provide a universal basic income for people. It would be a world where AI would be doing all our jobs and we would be sitting at home and focusing on building better things. Probably, the next Jarvis.
On the other hand, some people are already concerned about the threat of autonomous killer robots, citing Ukraine’s recent AI based drones that are automatically targeting and attacking targets without human control, calling them “killer robots”. Since these AI-driven drones might be able to perform better than the human-driven ones, people are expecting chaos, apart from just job displacement.
Science fiction no more – autonomous killer robots are here “Ukraine is using AI drones that can identify and attack targets without any human control, in the first battlefield use of autonomous weapons or ‘killer robots’” What happens next? “A year from now these drones will… https://t.co/hCLqxftY0L pic.twitter.com/GOI4fYlMDO
— AI Notkilleveryoneism Memes (@AISafetyMemes) October 14, 2023
On a funnier note, have you ever got mail from a Nigerian prince stuck in a foreign country who is asking you to send him money? If you thought these phishing emails were a problem, AI is going to give a major upgrade to it. Now it’s not going to be some guy sending these emails, but an AI agent, posing as a Nigerian prince.
The post OpenAI has Just Cracked AGI appeared first on Analytics India Magazine.
Jonty Rhodes was the highlight of the first day of Cypher. The South African player and legendary fielder, after retiring from playing the sport, is a coach for IPL teams. Rhodes was in high spirits as he reminisced about his time playing the sport while comparing it with how it has changed over the years.
“It is insane how the game has changed in most capacities. I remember Bob Woolmer was the first coach who used a laptop, he would flip it out and begin analysing the opposition.” he said. The beginnings of technology in cricket was a very basic exercise then, but now statistics and figures are used from a coaching perspective to how players are chosen, he explained.
Using technology in sports isn’t new. In cricket, which has more pauses in between the game as compared to football, for example, technology has calculated the statistics of every aspect of the game but Jonty said, “Cricket was never really driven by statistics until now, like when Virat Kohli comes up to bat, the opposition will employ a left arm bowler.” This is because of the statistics that show that Virat gets out against left arm spin more often than not. That’s what the sport has become now.
AI and Coaching
The analytics in cricket has moved beyond looking at individual players to finding patterns within the team. As a part of the coaching staff, he said, “We look at the power play, the middle section and the death overs whose patterns we look at separately and together. The strike rate, number of runs scored and the number of wickets etc are considered and then strategise our players.”
For example if a minimum score of 45-50 hasn’t been achieved in the powerplay with a loss of less than three wickets then it becomes incredibly difficult to make up for it in the middle section, he explained. For this the teams strategise based on different factors, the conditions of the day, who the opposition team is, what is the condition of the team collectively and each of the players etc.
“I think there is still a lot of potential in tapping personal player data,” Jonty said. There is a lot of data on the player’s physical exertion, from the watches and vests which helps not only in the training but also analysing the recovery rate of each player. “We especially saw that during the pandemic when players had to train in a bubble which is the worst way to train for a cricketer. And what’s interesting for me, I’m sort of seeing and reading technology that gives you neurological feedback of the player,” Rhodes said.
As a coach he says he is more interested in bringing out the best in the players for which it is essential to know where they stand. “The neurological feedback can change how we approach training the player. How to help them level their emotional ups and downs from the pressure of the game, is something we don’t have yet, but will take the game to the next level,” he said.
Allure of the game
Beyond the data and analytics the game is about the instinct of the players who have more experience. The coach doesn’t have to constantly check on the statistics of the players because a lot of factors are unknown. “When I first faced Inzamam-ul-Haq, I had never hit a wicket like that before but in that moment it felt right,” he said.
There is some fear of it becoming too analytical or dependent on technology that takes the charm away from the game. “When the surprise element pops up, which is quite often, what we depend upon is how prepared the team is. There are no substitutes for that,” he explained.
The evolution of data from the first IPL in 2008-2009 to now is exponential. “But the thing to remember is that we only needed a couple of pieces of advice or information to make a difference to how we practise or approach the game and that is true even now,” he said. Data and AI can’t take away the human element. “When new players who don’t have the stats and career yet are chosen, they’re not done so with an excel sheet but an instinct of their potential.” he wrapped up.
The post How Jonty Rhodes Coaches with AI appeared first on Analytics India Magazine.
The field of healthcare is increasingly attracting the efforts of the most prominent companies in artificial intelligence, with Microsoft being the latest example.
Last week, the company announced extensions to Fabric, the data analytics platform it unveiled in May, to enable Fabric to perform analysis on multiple types of healthcare data. Microsoft also announced new services in its Azure cloud computing service for, among other things, using large language models as medical assistants.
"We want to build that unified, multimodal data foundation in Fabric One Lake, where you can unify all these different modalities of data so that you can then reason over that data, run AI models and so on," said Umesh Rustogi, the general manager for Microsoft Cloud for Healthcare, in an interview with ZDNET.
The trend of multi-modality, which ZDNET explored in a feature article on AI this month, is increasingly important in healthcare, said Rustogi. "We have heard this from multiple customers, where they believe that if you combine multiple modalities of data that can unlock new insights, which is not possible by doing research just on one modality of data," said Rustogi.
Umesh Rustogi, the general manager for Microsoft Cloud for Healthcare.
Examples of such combined modalities include "simple things like building cohorts of patients based on criteria from their imaging results and their clinical results, [which] is one very common desired use case which is not very easy to do today," he said. Rustogi cited as exemplary of what some would like to do a 2020 study in the prestigious journal Nature. That article offers an overview of techniques for "data fusion" that can be "applied to combine medical imaging with EHR [electronic health records]."
Another one of the new Fabric capabilities is a "de-identification service," which uses machine learning forms of artificial intelligence to scrub clinical data to hide the patient identities in data such as doctors' notes. "It has been a very hard problem for the industry to solve as to how do you take those unstructured clinical notes, and then de-identify them in such a way that it's still meaningful for the research community," said Rustogi.
Rustogi's colleague, Hadas Bitran, head of Microsoft's Health AI and Health and Life Sciences, discussed several new offerings for AI from the Azure web services business.
The Azure AI Health Insights offering consists of pre-built machine learning AI models. Three models are initially being offered in a preview stage:
Patient timeline, which "uses generative AI to extract key events from unstructured data, such as medications, diagnosis and procedures, and organizes them chronologically to give clinicians a more accurate view of a patient's medical history to better inform care plans";
Clinical report simplification, which "uses generative AI to give clinicians the ability to take medical jargon and convert it into simple language while preserving the full essence of the clinical information so that it can be shared with others, including patients";
Radiology insights, which "provides quality checks through feedback on errors and inconsistencies. The model also identifies follow-up recommendations and clinical findings within clinical documentation with measurements (sizes) documented by the radiologist."
Those three models are being added to several pre-built models that were already offered for clinical trials matching and for oncology phenotype-based models.
A new offering called Azure AI Health Bot uses large language model technology to retrieve answers to medical questions from sources including a healthcare organization's own database, or the US National Institutes of Health and the US Food and Drug Administration.
"An idea here is that this service helps customers create specialized co-pilot experiences," Bitran told ZDNET in the same interview with Rustogi.
"What's also interesting about this is that you are able to do a cascading effect," said Bitran. "So, use your own sources, and if there's nothing in your own sources, you are able to also provide answers based on the credible sources, and then, if there's nothing in the credible sources, then you can just fall back to a generic answer."
Of course, there's a lot of skepticism at present about using generative forms of AI, such as large language models, in sensitive practices such as healthcare. How does Microsoft think about such concerns?
Hadas Bitran, head of Microsoft's Health AI and Health and Life Sciences.
"That's a very good question, and a very relevant one," said Bitran. "I definitely share a view that large language models need something in addition to them in order to provide good results."
"The way we're approaching it, is, for every model that we create, if we're using large language models, they will always be accompanied by healthcare-specific safeguards," said Bitran.
"One of the more interesting approaches [to safeguards] is using smaller models, and rule-based models, in a hybrid model with the LLM, to keep the LLM honest, if you will," said Bitran.
For example, in the pre-built model for clinical report simplification, "we don't just ask the language model to explain it to me; we're also implementing a lot of pre-processing and post-processing logic that allows us to take the outcome of the simplification, measure it according to simplification performance metrics," explained Bitran. "We then apply some cross-reference to it to see whether the results are actually a simplification of the source, or whether there are all sorts of fabrications or things that are missing."
Bitran noted that the work on healthcare is done within what Microsoft has outlined as its "responsible AI framework," which continues to be evaluated.
"That responsible AI framework is not just about privacy and security and accessibility and transparency, etc," said Bitran. "It is also about correctness and accountability and about fairness."
"Last but not least, our models are not intended to replace the physician," said Bitran. "There is always a human individual; they're intended to equip the clinicians with tools that would alleviate the burden, that would help them in their work."
The team behind Opera understands that users want to be able to work (and play) efficiently. In the modern area of burgeoning AI, this means users want to be able to work with features like Aria and integrate them more and more into their daily lives.
Also: How to use Opera's built-in AI chatbot (and why you should)
According to Joanna Czajka, product director at Opera, "This year, we redesigned our browser to make Opera users' interaction with AI better. With this release, we are improving their experience in places where it matters the most: Queries and content creation." She adds, "What people want is to create the content they need or get the best possible query results as fast, as accurately, and with as little effort as possible. With this update, we are making that happen."
The new Refiner tool
To that end, the Opera team has added a couple of new features that will go a long way to simplifying your Aria AI interactions.
The first new feature is called Refiner. This tool helps Aria refine part of an answer it returns. If you're not satisfied with a response, you only need to select the part of the response you're unhappy with and then click Rephrase.
Also: How Opera's search pop-up can save you time and clicks
Aria will then keep repeating the answer to your prompt, reworking the selection you made (while keeping everything else the same).
Easier composing
The improved Compose feature allows you to refine your prompts with handy graphical elements. With this tool, you can configure the following:
Tasks: Select the type of content you want to write (such as a blog post, email, essay, presentation, and more)
Description: Allows you to provide a topic and a bit more context
Tone: Allows you to select from formal, informal, neutral, academic, business, or funny
My Style: Allows you to refine your style
Length: Allows you to select from short, medium, or long
The My Style feature should be appealing to many users. It uses AI to train on your style of writing and will ask you to write the following:
A formal letter of complaint
A product review
A casual text message
Once you've done this, Aria will train on your style and will then be better capable of outputting content in sync with your personal style of writing.
Also: Firefox vs Opera: Which web browser is best for you?
As of writing this, I've yet to see the Refiner tool show up in either Linux or MacOS versions of Opera (I'm running version 103.0.4928.26 on both). However, the prompt refining and My Style features are already available.
It seems that nearly every major company is finding ways to incorporate AI into their business plans, investing in building or adopting generative AI models to carry out specific tasks. A new report forecasts that the spending will only increase.
International Data Corporation (IDC) predicts that by 2027, spending on generative AI solutions, which includes software, related infrastructure hardware, and IT/business services needed to implement the generative AI, will reach $143 billion.
Also: 80% of enterprises will have incorporated AI by 2026, according to a Gartner report
The IDC forecast sees enterprises investing nearly $16 billion worldwide in these generative AI solutions in 2023, representing a compound annual growth rate (CAGR) of 73.3% over the 2023-2027 forecast period.
"Generative AI is more than a fleeting trend or mere hype. It is a transformative technology with far-reaching implications and business impact," said Ritu Jyoti, group vice president at IDC.
For comparison, the growth of generative AI investments is more than twice the growth rate in overall AI spending and 13 times greater than the CAGR for worldwide IT spending over the same period, according to IDC.
IDC forecasts that the investments will follow a natural progression over the next several years following organizations' shifts from early experimentation to aggressive build-out to widespread adoption.
Also: Mass adoption of generative AI tools is derailing one very important factor
"The rate of GenAI spending will be somewhat constrained through 2025 due to turbulence in workload shifts and resource allocation, not just in silicon but also in networking, facilities, model confidence, and AI skills," noted Rick Villars, group vice president at IDC.
By the end of the forecast period, IDC predicts that generative AI spending will account for 28.1% of overall AI spending, a significant increase from 2023, in which generative AI spending accounted for only 9%.
AI for Everyone: UiPath’s Autopilot Levels the Playing Field in Business Automation October 16, 2023 by Drew Jolly
AI, particularly generative AI, is poised to play a crucial role in future business strategies, potentially introducing new product or service offerings, enabling data monetization, and allowing for increased personalization in offerings. Enterprise automation software company UiPath wants to lead this transformative wave with its new Autopilot feature, aiming to integrate AI capabilities into everyday operations, reshaping industries and bridging the gap between technology and actionable business outcomes.
UiPath introduced Autopilot last week at its global user conference FORWARD VI. Promising to transform paper documents into automation-powered apps, UiPath says that Autopilot integrates Generative AI, Specialized AI, and automation, targeting a broad audience, from developers to business analysts.
The company says the new features aim to integrate AI-specific copilots across apps and contexts while focusing on bringing AI closer to daily work through automation. With an emphasis on user-friendly interfaces and natural language capabilities, UiPath says Autopilot can enable even those without technical backgrounds to harness the power of AI in their work processes.
By making it accessible to a wider range of users, UiPath is hoping to democratize the AI landscape. Graham Sheldon, Chief Product Officer at UiPath, commented on the launch, noting that Autopilot is designed to act as a compass, guiding users toward the next best actions, whether that's interacting with a customer or handling internal operations. The goal is clear: to make AI not just a lofty concept but an integral, tangible part of how businesses operate daily.
And as AI becomes more ubiquitous in the modern business landscape, companies need to be proactive in embedding it into their core processes. UiPath is lining up Autopilot to be a practical avenue for businesses to truly embrace the potentials of AI.
Credit: UiPath
By bridging the divide between complex AI tech and everyday business tasks, UiPath is not only setting the stage for the next phase of digital transformation but is also emphasizing the importance of adaptability and inclusivity in the AI ecosystem. As the lines between AI capabilities and human potential continue to blur, forward-thinking tools like Autopilot could ensure that businesses remain ahead of the curve, harnessing the full power of AI.
And a recent study conducted by UiPath and Bain Capital reveals a notable trend: a majority of executives surveyed have already incorporated some form of Generative AI into their business operations.
“Businesses must go beyond deploying this technology and fundamentally rethink and redesign business models to integrate AI and automation seamlessly,” said Bain's Ted Shelton in a recent press release. “The truly future-proofed organizations will be agile, with a fluid culture and design, constantly evolving, and reconfiguring in tandem with technological advancements in AI.”
In an era where AI is rapidly permeating most business operations, UiPath's introduction of Autopilot could serve as a beacon for transformative technological integration. By combining Generative AI, Specialized AI, and automation into a user-centric interface, the company is taking significant strides in AI democratization.
But It's not merely about deploying AI. The crux will lie in seamlessly integrating it into business models, ensuring agility and adaptability. With tools like UiPath’s Autopilot, the future of business lies in an integrated, AI-driven ecosystem where tech and human potential coalesce to redefine how work gets done.
The applications of artificial intelligence in health care are numerous. But they are largely dominated by older AI technology; newer things such as so-called generative AI and large language models (LLMs) are the craze of the moment, but they are deemed too risky to be used to any great extent in health care given the sensitive nature of health applications, as ZDNET has recently reported.
Efforts in open-source software could help advance generative AI by making it a little bit easier to look inside the "black box" of AI compared to closed programs such as OpenAI's ChatGPT.
Also: How does ChatGPT actually work?
In that spirit, AI computer maker Cerebras Systems last week announced a joint effort with partner M42, an operator of healthcare facilities in 27 countries, to offer an open-source LLM designed for health applications, to serve as an "assistant" to healthcare providers.
The program, called Med42, is a refinement of Llama 2, the open-source LLM released by Meta Properties this year, using a special health-related data set compiled by the companies.
"It's blazing a trail in the use of AI in delivering healthcare," said Cerebras co-founder and CEO Andrew Feldman in an interview with ZDNET.
Applications of the program as an assistant to physicians include medical question answering, patient record summarization, aiding medical diagnosis, and general health Q&A, according to the companies. It does not include training physicians, they emphasized.
The Med42 program uses the 70-billion-parameter version of Llama 2. The work of fine-tuning was performed by Cerebras and M42 in conjunction with Core42, a managed services and IT firm that does fundamental AI research. Both M42 and Core42 are owned by Cerebras customer G42, a global conglomerate.
The Med42 neural net was fine-tuned with a data set of 700,000 question-and-answer pairs from publicly available sources, "curated by M42 and reviewed by our team of medical experts," said M42 in an email to ZDNET. "The dataset included multiple choice questions, medical flashcards, among others," it said.
"Med42 has not been trained using patient data or personally identifiable information," said M42.
The M42 code is available now on HuggingFace, along with performance data.The companies plan to release enhancements as they "refine and test the model collaboratively" with health care professionals "to help enhance its capability and performance." When asked if the data set itself will be released, the companies told ZDNET in an email, "This is still to be determined."
Also: Microsoft unveils extensions to Fabric, Azure for healthcare AI
The fine-tuning was done on Condor Galaxy, a massive AI computer that Cerebras built for G42 this year, which Cerebras calls "the world's largest supercomputer for AI." According to Cerebras, "Rapid setup and reduced training time were made possible by the 82 terabytes of memory and the 54 million AI cores in the 64 Cerebras CS-2 systems inside of CG-1."
"What you have are all these interesting applications being run on top of Condor Galaxy, and that's unlike any other startup's hardware," said Feldman. "We're really moving the industry forward."
Feldman noted in a follow-up email that "All parameters [of Llama 2] were fine-tuned, and this was made possible by the vast memory available on Condor Galaxy 1 […] The setup and training for 3 Epochs was accomplished in 5 days, which would have taken months on a large cluster of GPUs."
In the performance data on HuggingFace, the companies note that "Med42 achieves competitive performance on various medical benchmarks, including MedQA, MedMCQA, PubMedQA, HeadQA, and Measuring Massive Multitask Language Understanding (MMLU) clinical topics."
Also: What is HuggingChat? Everything to know about this open-source AI chatbot
On the US Medical Licensing Examination, or, USMLE, sample exam, the program "achieves a 72% accuracy," according to M42, "surpassing the prior state of the art among openly available medical LLMs." It also surpassed by a wide margin OpenAI's closed-source GPT 3.5, which garnered 59.6% accuracy, though Med42 fell short of GPT4's 84.3% accuracy.
"You take a very big pre-trained model like Llama 2 70 billion, and if you bring to it really interesting data sets, pioneering data sets, you can have them do really interesting things, and done at a fraction of the time and the power draw of something like GPT 3.5," said Feldman.
Cerebras has been especially active in open-source projects of late. In March, the company published as open-source several versions of generative AI programs to use without restriction.
In August, the company unveiled the world's most powerful Arabic-language LLM, Jais-Chat, as an open-source program.
For the moment, Med42 is not in production. "Following successful testing, Med42 will be made available for clinical deployment," the companies said in an email to ZDNET.
"Importantly, Med42 will have the capability of being deployed on-premise, fully customized to healthcare providers' needs, using owned data sources and limiting the ability for external intrusions," they added. "We are prioritizing safe application of the technology over speed to production and are committed to extensive safety evaluation of the model before rolling it out."
Australian software giant Atlassian recently announced its plans to acquire video messaging platform Loom for nearly $975 million.
In the official press release, the duo announced that their joint investments in AI would enhance asynchronous video capabilities. Their users will be able to move between recorded clips, transcripts, summaries documents and workflows. The goal is to facilitate collaboration among individuals, irrespective of their geographical locations and time zone differences.
“Async video is the next evolution of team collaboration, and teaming up with Loom helps distributed teams communicate in deeply human ways,” Mike Cannon-Brookes, co-founder and co-CEO of Atlassian, stated while announcing the deal.
The acquisition news has not been well received as Loom’s valuation has dropped post pandemic, whereas Atlassian still sees its 25 million customers, and more than 5 million video conversations per month, as a valuable asset.
Back in 2016, Loom emerged as a pioneering web-based tool designed to capture a user’s facial expressions and on-screen activities, simultaneously. Over time, it was widely adopted, particularly among companies with dispersed workforces. Loom’s primary innovation lay in facilitating asynchronous video communication, enabling individuals to message their peers, who could then respond at their own convenience, thereby offering an alternative to real-time meetings and conferences.
Loom-ing Over
Loom’s tools let users record their screens, camera and microphone to make and share videos. The San Francisco-based company boasts Sequoia, Kleiner Perkins and a16z among its investors but has not had a good tiding monetarily in the recent past.
The startup had raised nearly $204 million since 2016 but as times changed, so did the company’s value. Loom’s co-founder Vinay Hiremath said that Atlassian has been a long and true believer of Loom.
“To put the longevity and growth of the relationship into perspective, Atlassian was all-in on Loom when we were doing just under 1 million looms per month in 2019. We are now doing over 7 million looms recorded per month,” he said on X.
He said that this “acquisition has come at a time when Loom’s financials and growth are very strong”. But last June, the company had let go 14% of its employees a year after it joined the unicorn club.
Loom’s customer list reads like a who’s who of corporations across a variety of verticals, including Ford, Tesla, Disney, Walmart, Goldman Sachs and Amazon, to name but a few.
But with a free tier to push the business, perhaps too many users were free and not enough were in the paying category. It seems that the writing might have been on the wall last June when the company announced the layoff.
The latest deal with Atlassian is expected to be completed in the fiscal years ending June 2024 and 2025 but is unlikely to wake up the sleepy M&A market.
Atlassian has a record of failed acquisitions including Bitbucket, HipChat, and OpsGenie. While Loom has been an industry favourite, almost a billion being spent seems more for the company’s existing user base than the tech behind the tool.
On a brighter note
However, the partnership suggests a promising outlook for both companies, despite their individual struggles to establish a presence. Hiremath affirmed that the company will continue to ship new product value to the platform besides integrating across Atlassian’s products.
“The distribution Atlassian can provide Loom and vice versa is going to create a lot of customer value that would take much longer to create solo,” he said on X. He further stated that Loom’s goal is to make this one of the best software acquisitions in history.
Hiremath is optimistic about the deal since he wants people to look back on this moment and say “Atlassian got a steal” and continue to say “Loom just keeps getting better” and “How could I go back to Confluence/JIRA/etc. without video embedded into it?”.
The post Why Atlassian is Acquiring Loom appeared first on Analytics India Magazine.
Generative AI is being combined with DevOps systems that can both pinpoint causes of application failure and predict failure.
It's too soon to trust Microsoft's GitHub Copilot to automatically fix your programming code. Microsoft itself has said that the program, sold as a $10 per month add-on to GitHub, "does not write perfect code," and "may contain insecure coding patterns, bugs, or references to outdated APIs or idioms."
The dream of automation, however, suggests that someday, artificial intelligence will predict a fault in a program that can break functionality, or bring systems down, and not only warn a developer before the code goes into production, but also tell them how to alter code to avert the problem. AI might even be able to reach into the application code and automatically fix it for the programmer, saving them significant effort.
The makings of such a future can be seen in today's tools for DevOps and observability. DevOps tool maker Dynatrace has for a number of years been building what it calls "causal AI," and "predictive AI," to identify why programs go down, and to predict how they'll fail.
Also: AI will change software development in massive ways
The next stage is wrapping generative AI around those observability tools to give coders suggestions as to how their code is going to run into trouble and how to alleviate it.
"The typical request from a CIO is, please fix my system before it actually fails," says Bernd Greifeneder, chief technology officer of Dynatrace.
"The typical request from a CIO is, please fix my system before it actually fails," said Bernd Greifeneder, chief technology officer and co-founder of Dynatrace, in an interview with ZDNET. Dynatrace is a commercial software vendor in the DevOps and Observability market that sells tools for application lifecycle management.
Consider an everyday systems pitfall: running out of disk space in Amazon's AWS.
"It's totally ironic," noted Greifeneder. "Even in these days of super-high tech, it is a problem that cloud disks somewhere at AWS run out of disk space, and we have to trigger API calls in order to resize them. We don't want to resize them [the disks] up-front because it's costly, so we want to optimize what we use, but the usage patterns can change depending on how many customers we have in our clusters and so forth."
What's needed is to create code that will spring into action when an out-of-disk error looks likely based on past performance.
To tackle the problem, the company first identifies a "root cause" of a disk failure with the combination of causal and predictive AI. These two tools are not based on large language models and other generative AI. Instead, they rely on older, more well-established forms of artificial intelligence that can be counted on to produce rigorous, consistent results.
In the case of causal AI, the program employs several algorithms including quantile regression, density estimation, and what's known as a random surfing model. Unlike neural nets that are trained on a static set of data to detect correlations between data points, the causal programs are used to traverse a graph representing elements of a company's IT system and their relationships.
"Typical statistical models or neural network type of learning models do not work for dynamic IT systems in the bigger scope," said Greifeneder, because variables change too much. "Our customers may have tens of thousands to hundreds of thousands of pods, and many of them are interconnected, and change while traffic is routed, and things scale, and there are different versions, etc."
To build what Greifeneder calls an "in-memory, real-time model" of a customer's entire IT system, the causal AI programs construct a "multi-dimensional model that has the causal, directed dependency, sort of like a multidimensional graph" of all the entities — from what cloud service it is to what version of Kubernetes is being used to what app is running. That model, called Smartscape, is consulted whenever there is a system issue that raises alarms, "inferring the root cause based on traversing that Smartscape model."
Also: Businesses need pricing clarity as generative AI services hit the market
That causal model won't anticipate variations in the business, however. "It knows the root cause" of things, "but what it does not know, is, what is your business pattern," said Greifeneder, meaning, things such as, "Monday morning at 8:00 AM, you have a big spike in usage for whatever reason."
For such aberrations, "there needs to be some form of history-based learning," said Greifeneder.
To achieve that historical learning, a predictive AI component uses another set of well-developed tools, such as an autoregressive integrated moving average, which is an algorithm that's particularly attuned to piecing together patterns occurring in data over time.
Dynatrace's casual and predictive AI use a variety of time-tested algorithmic approaches to analyze signals of what is going on in a production application.
Crucially, predictive AI does not look only at back-end systems, such as the server. It also receives signals from endpoints in a network, things such as how the end user is experiencing lag or interrupted service.
"Looking at server-side systems alone is not good enough," said Greifeneder. "Real user monitoring, for instance, or API service monitoring, is an important aspect in understanding the dependencies."
While a CIO cares most about systems, user issues may crop up even when servers are running fine, so back-end and user experience both need to be measured and compared.
Also: What analysts are saying about the future of generative AI
"Sometimes we meet the IT-only person who cares only about their servers — 'Oh, my server is up' — but, actually, users are frustrated," he said. "The opposite exists: Just because one of those CPUs goes wild, it doesn't mean the end user is impacted."
Returning to the disk space example, the causal and the predictive AI can anticipate a future disk issue. "We can extrapolate from the past days and weeks of usage in the cluster to see, 'Oh, we run the risk that in a week from now, we might run out of disk space,' " said Greifeneder.
That is the impetus to take proactive steps, such as, "Let's trigger now a workflow action from Dynatrace's automation engine to call an API into AWS to resize the disk and therefore automatically prevent an outage that we had in the past because of this."
It's here that generative AI gets looped into the process. The Dynatrace umbrella program, Davis AI, this year added a component called Davis CoPilot that rides on top of the causal and predictive systems.
Also: 4 ways to increase the usability of AI
A user can type to the CoPilot, "create me an automation that actually prevents this [disk outage proactively]." The CoPilot can send an inquiry to the causal and predictive AI to ask what disks are being referred to in that prompt. In response, the Davis program uses the Smartscape and the predictive information to create a prompt with all the contextual details that are required to understand the IT system in its current state.
That prompt is then sent to the CoPilot, which, once given the details, "will give you back the template of the workflow to automate" the disk re-sizing, explained Greifeneder. "It will give you, as the user, the ability to review and say, OK, this is approximately right, thank you, you helped me get 90% there," which can save the systems engineer time versus building a workflow from scratch.
The next step is for the Davis AI program to bring all these observations back to the programmer at the time they are first coding the application. The holy grail of application development is to prevent coding that causes faults before that application is put into production rather than having to fix things later.
One approach is what Dynatrace calls a guardian. A DevOps individual can ask the CoPilot in natural language to create a guardian to watch over a particular application performance goal before that application is put into production. The company terms this "defining a quality objective in the code." The causal and predictive elements are then used to verify whether or not the code will meet the objectives that have been defined.
Also: How I tricked ChatGPT into telling me lies
Of course, if the Davis AI notes a potentially problematic code, the issue is then how to fix it. It is possible to have the Davis CoPilot advise the programmer on possible code fixes, though that is still an emerging area.
"We are thinking about, with this Davis CoPilot, providing recommendations on how we identified this vulnerability in production based on your technology stack, and Davis CoPilot provides you these recommendations that you should check out to fix in your code," Greifeneder told ZDNET.
It's still early in the use of generative AI for those kinds of code fix recommendations, said Greifeneder. While the causal-predictive AI is engineered to be reliable, the generative algorithms still suffer from the phenomenon of "hallucinations," meaning that the program confidently asserts inaccurate information.
"What is reliable is what comes from the causal AI because that is the accurate system state," he said. "So, we know exactly what's there; what is not reliable is the potential recommendation on how to change the code because this comes from the public GPT-4 models."
Also: Everyone loves low-code/no-code development, but not all are ready for it
Therefore, code suggestions for remediation may start from a valid premise, but they run into the same issue as GitHub Co-pilot: not having a really rigorous sense of what code is appropriate. There's a need to integrate large language models more closely with the tools that Dynatrace and others provide, to give some grounding to generative AI's suggestions.
Formal studies of GPT-4 and their ilk report very mixed results in finding and fixing code vulnerabilities. The technical paper released by OpenAI with the introduction of GPT-4 in March cautioned against relying on the program. GPT-4, it said, "[…] performed poorly at building exploits for the vulnerabilities that were identified."
A study in February of GPT-4's predecessor, GPT-3, by University of Pennsylvania researcher Chris Koch, was encouraging. It showed that GPT-3 was able to find 213 vulnerabilities in a collection of GitHub repository files curated for their known vulnerabilities. That number was well above the 99 errors found by a popular code evaluation tool named Snyk, a form of "Static Application Security Testing," or SAST, commonly used to test for software vulnerabilities.
But, noted Koch, both GPT-3 and Snyk missed a lot of vulnerabilities — they had a lot of "false negatives," as they're known.
A subsequent study, by cybersecurity firm PeopleTec, built upon Koch's work, testing an updated GPT-4 released in August. It found that GPT-4 uncovered four times as many vulnerabilities in the same files.
However, in both studies, GPT-4 was tested on files representing a grand total of just over 2,000 lines of code. That is minuscule compared to full production applications, which can contain hundreds of thousands to millions of lines of code, across numerous linked files. It's not clear that successes on the toy problems of the GitHub files will scale to such complexity.
Also: ChatGPT answers more than half of software engineering questions incorrectly
The race is underway to try and amplify language models for that greater task. In addition to Dynatrace, privately held Snyk Ltd. of the UK, which sells a commercial version of the open-source tool, offers what it calls "DeepCode AI." That technology, Snyk claimed, can avoid the stumbles of generative AI by integrating it with other tools. "DeepCode AI's hybrid approach uses multiple models and security-specific training sets for one purpose — to secure applications," the company stated.
It's clear that generative AI has a ways to go to solve even simple kinds of programming debugging and fixing, leaving aside the complexity of a live production IT environment. The great shift left via AI is not here yet.
What is on the horizon, with Davis Copilot and efforts like it, is using generative AI as a new interface to help coders examine their own code more aggressively both before and after they ship that code.
ByteDance’s video editor CapCut targets businesses with AI ad scripts and AI-generated presenters Sarah Perez @sarahintampa / 9 hours
CapCut, the ByteDance-owned video editing app that’s the company’s second to hit $100 million in consumer spending after TikTok, is now expanding into business tools. Known today for its easy-to-use templates, tight integration with TikTok, and rapid adoption of AI effects and filters, CapCut has been a top consumer video editing app that now regularly ranks in the top 10 or 20 Overall apps in the iOS App Store. Now the company will bring its set of tools to advertisers and creators with the introduction of CapCut for Business.
This business-focused extension of the CapCut platform prioritizes tools that help marketers, brands, small businesses, and creators generate ads and branded content. These tools will be made available across the CapCut app for desktop, mobile, and tablets, the company says.
Image Credits: CapCut
Included in the offering is an AI-powered script generation tool that helps advertisers come up with script ideas based on their product or business description, as well as thousands of commercially licensed business templates, a smart tool for converting URLs of product or landing pages into videos, and more.
AI plays a heavy role in the new service as well, beyond ad script generation.
CapCut for Business also allows marketers to access AI-generated presenters who can show off a company’s products through demos and explainer videos. And a virtual try-on feature uses AI models to allow customers to virtually try on products and generate photos that showcase the product. This latter option is aimed at e-commerce businesses and clothing merchants.
Image Credits: CapCut
Unlike the consumer app, CapCut for Business is designed for use across teams. Collaboration features let users work on their ads with other team members, agencies, and creators, where anyone can be given permission to edit, access, review, or add notes and suggestions.
While the tools could help businesses craft videos to advertise their products across TikTok and other short-form video platforms, they can also be used for organic content — like videos posted to the brand’s TikTok account, for example.
Image Credits: CapCut
Originally launched in China in 2019 as JianYing, its sister app for non-China markets, CapCut, expanded globally across the following years and grew in popularity thanks to its ties to TikTok. As of August 2023, the app was used by 490 million iOS and Android users worldwide, reports market intelligence provider data.ai — that equates to roughly 25% of TikTok’s user base. Top markets outside China include the U.S., U.K., Germany, Egypt, Saudi Arabia, Mexico and Brazil, which have seen sizable increases in downloads over the first half of this year. In total, CapCut reaches 170 countries outside China.
CapCut also surpassed Splice to become the most profitable video editing app globally during the first half of 2023, pulling in a record high of $50 million, making it ByteDance’s second app to top $100 million globally. Data.ai attributes the increased adoption and revenue gains to the more recent introductions of AI features, like templates, effects, and filters. After adding generative AI templates and effects this April, for instance, CapCut saw a download spike followed by continuous revenue increases, the firm said. ByteDance also offers a CapCut plugin for ChatGPT users.
Now the company is positioning its editing app as a way for consumers to make compelling videos for social media, including TikTok, and for marketers to easily do so as well, without having to spend heavily on advanced video editing software. In selected case studies, CapCut reports business users were able to increase their video output, views, and engagement, sometimes by triple-digit percentages, and were able to reduce the average CPM in other cases and drive online sales. Broader real-world results will not be available until the software gets into the hands of more marketers, of course.
The new CapCut for Business software is available to brands, marketers, and creators starting today at no cost.
pic.twitter.com/NwZW7YZiH5
(@AISafetyMemes) October 14, 2023
