Meta has announced the release of four new AI models and additional research artifacts at Meta FAIR, as part of its commitment to fostering an open ecosystem. These releases aim to inspire innovation in the community and advance AI in a responsible way.
Today is a good day for open science. As part of our continued commitment to the growth and development of an open ecosystem, today at Meta FAIR we’re announcing four new publicly available AI models and additional research artifacts to inspire innovation in the community and… pic.twitter.com/8PVczc0tNV
— AI at Meta (@AIatMeta) June 18, 2024
The new AI models include Meta Chameleon, which offers 7B and 34B language models supporting mixed-modal input and text-only outputs.
Additionally, Meta Multi-Token Prediction is a pre-trained language model designed for code completion using multi-token prediction.Using this approach, Meta trains language models to predict multiple future words simultaneously, rather than one at a time. This method enhances model capabilities, improves training efficiency, and allows for faster speeds.
Meta JASCO, another new release, is a generative text-to-music model that accepts various conditioning inputs for greater controllability. The accompanying paper is available today, with a pretrained model to be released soon.
Meta AudioSeal is an audio watermarking model designed specifically for the localised detection of AI-generated speech and is available under a commercial license.
Alongside these models, Meta is releasing additional Responsible AI (RAI) artifacts, which include research, data, and code aimed at measuring and improving the representation of geographical and cultural preferences and diversity in AI systems.
Meta emphasises that access to state-of-the-art AI should be available to everyone, not just a few Big Tech companies. The company is eager to see how the community will utilise these technologies.
The Dreame X40 Ultra is available for $1,900, with a $120-off coupon available on Amazon now.
The X40 Ultra is a high-performing robot with excellent mapping capabilities and strong 12,000Pa suction. It performs exceedingly well on carpet and hard floors, with great object avoidance and high customization power.
Though it can recognize and show snapshots of obstacles, it sometimes gets tangled in cords and loses one or both mop pads. I also found some connectivity issues with the app.
As a robot vacuum and mop fanatic, I always look for the next big thing in home cleaning robotics. The Dreame X40 Ultra may just be it.
Also: 5 things I do to keep my robot vacuum in good health (that you might be forgetting)
This robot vacuum and mop combination has so many features that keeping up with them during testing was challenging. Unboxing the Dreame X40 Ultra was a joy, mainly because the robot looks nicer than most eyesores. The base station features a grooved front and gold accents, while the robot looks sleek and simplistic.
View at Amazon
Setup was quite a breeze — you download the Dreamehome app and follow the instructions to add it to your Wi-Fi and set up your preferences. After setting it up, I found some issues with the app connecting to the Dreame X40 Ultra, making me wait a few minutes or force-quit the Dreamehome app to relaunch it. A month later, I found this still happens occasionally, which is disappointing as I don't have that issue with other robots.
The robot features market-leading suction power at 12,000Pa, making it excellent for carpeted homes or homes with pets. It's efficient and fast, easily cleaning across my floors and navigating obstacles while staying true to its map.
One of my two favorite features of the Dreame X40 Ultra robot is its great object avoidance feature. This is one of two robots in my home that doesn't require me to pick up every object from the floor before running it. This means I can send it out to clean on a schedule or when I'm away from home and rely on it to consistently deliver a clean home without getting stuck on a kid's sock under the coffee table.
My second favorite thing about it is that it has magnetic mop pads that not only lift to avoid getting carpets wet but can also be set up in the app to have the robot leave the mop pads at the base, vacuum the carpets first, and then vacuum everything else. Then, it returns to the base to reattach its mop pads and mop the entire house, avoiding the already clean carpets.
The X40 Ultra's mop pads lift to about 10.9 mm, not enough to keep my living room's medium pile carpet dry. This means I have to set that rug as a no-go zone for robot vacuum and mop combinations like the Yeedi M12 PRO+, which lifts its mop pads to 9 mm. The X40 Ultra's ability to detach its mop pads at the base station and clean carpets first is a game changer. The app also lets you enable intensive carpet cleaning, where the robot slows down and cleans carpets twice by cross-walking.
The Dreamehome's app has so many customizations that it reminds me of everything you can do with the Roborock S8 MaxV Ultra, a direct high-end competitor with many of the same features. Like the S8 MaxV Ultra, the X40 Ultra takes photos of obstacles and can take photos of your pet in passing if it spots them.
Dreame uses AI for visual recognition to identify and add objects to your map. Of course, this isn't always accurate, which is why my toddler was mistaken for a pet, but it's pretty entertaining and useful that the robot lets you see the obstacles it finds in its path. You can also drop into the camera's feed, seeing what the robot sees using your smartphone.
The X40 Ultra's camera can detect stains on hard floors and rev up the robot's mopping power to scrub up the stains. When this happens, the side brush automatically lifts to avoid getting wet or spreading wet messes. When the built-in turbidity sensor detects too much dirty water, the robot returns to the base station to rewash its mop pads and resumes its cleaning session.
So, how is this robot not perfect? To address the elephant in the room, the Dreame X40 Ultra is one of the most expensive robot vacuum and mop combinations I've seen, at $1900. The Roborock S8 MaxV Ultra is priced at $1800, and I already find that to be steep. Spending almost two grand on a robot that will roll around your dirty floors isn't an easy purchase.
It's also simply not perfect because nothing is. I found that the X40 Ultra often tries to go over some extension cords rather than avoid them, which almost always results in one or both of its mop pads coming off and the robot getting stuck without them. This is a bigger inconvenience when I've left it alone to clean the house and come back to see it surrounded by dirty floors in a corner.
ZDNET's buying advice
When it comes down to it, the Dreame X40 Ultra is the smartest robot vacuum and mop I've ever tested.
One simple example: If you've ever had robot vacuums, you've likely experienced them aimlessly roaming around when it's time to return to the dock, only to pause two feet away to say it can't find the charging station.
This Dreame X40 Ultra never does that. As a robot vacuum reviewer, I decided to bring out 11 robot vacuums in a group to take photos of them. After the photos, the X40 Ultra accidentally got bumped when someone tried to walk over the robot labyrinth and began returning to the dock. After expertly navigating through 10 of its pals, weaving to and fro like a cyclist through a traffic jam, it went straight to the charging dock and began charging. I tried this with three other robots, but none found their charging docks.
The fact that I have three little kids making as many messes as summer break allows, and I don't have to worry about this robot getting its roller brush stuck makes the Dreame X40 Ultra one of the best robot vacuum and mops I've ever tested.
OpenAI’s chief scientist Jakub Pachocki’s X account was recently hacked, according to Bloomberg. The hacked account posted that OpenAI is bridging the gap between AI and blockchain and claimed that all OpenAI users are eligible to receive a portion of a new token supply.
Someone has gained access to Jakub Pachocki's X account (OpenAI Chief Scientist). This comes one year after Mira Murati's account was compromised (OpenAI CTO). pic.twitter.com/qXZf9QuTjh
— Smoke-away (@SmokeAwayyy) June 19, 2024
“We’re very happy to announce $OPENAI: the token bridging the gap between AI and blockchain technology. All OpenAI users are eligible to claim a piece of $OPENAI’s initial supply. Holding $OPENAI will grant access to all of our future beta programs,” read the post on X.
Bloomberg confirmed that Pachocki’s account was compromised. “OpenAI is not getting into crypto, company comms just confirmed with me. Jakub Pachocki’s tweet that is making the rounds appears to be a hack,” a Bloomberg journalist posted on X.
OpenAI is *not* getting into crypto, company comms just confirmed with me. Jakub Pachocki’s tweet that is making the rounds appears to be a hack.
— Shirin Ghaffary (@shiringhaffary) June 19, 2024
Jakub Pachocki recently took over as the company’s chief scientist, succeeding Ilya Sutskever, who departed to pursue personal endeavors. As chief scientist, Pachochki leads OpenAI’s research efforts, focusing on scaling deep learning systems and advancing AI research.
Pachocki joined OpenAI in 2017 and has held several key positions, including Research Lead for the Dota team, leader of the Reasoning Team, and head of the Science of Deep Learning team. He also served as Director of Research, leading the development of GPT-4 and OpenAI Five and conducting fundamental research in large-scale reinforcement learning and deep learning optimisation.
During last week's introduction of Apple Intelligence — Apple's artificial intelligence initiative — software engineering head Craig Federighi announced that the company will run some generative AI models in a secure cloud computing environment when the models require extra horsepower.
Called Private Cloud Compute (PCC), the service will be subject to scrutiny by outside security experts. Said Federighi: "Just like your iPhone, independent experts can inspect the code that runs on these servers to verify this privacy promise." The point: To verify certain privacy promises by Apple, including that user data will never be stored on PCC servers, but will be expunged from memory once a request is fulfilled.
Federighi did not go into detail about how the PCC servers will be inspected or audited by security researchers, but a subsequent blog post by Apple technical teams states the PCC servers will run a distinct version of the company's operating system software that researchers will be allowed to inspect.
"When we launch Private Cloud Compute, we'll take the extraordinary step of making software images of every production build of PCC publicly available for security research," states a post by the Apple Security Engineering and Architecture and collaborating teams.
The article goes on to say that Apple will "periodically also publish a subset of the security-critical PCC source code, [and] in a first for any Apple platform, PCC images will include the sepOS firmware and the iBoot bootloader in plaintext, making it easier than ever for researchers to study these critical components."
Apple emphasizes that its devices "will be willing to send data only to PCC nodes that can cryptographically attest to running publicly listed software" as a means to ensure its privacy and security guarantees.
Apple makes various promises about the safety and security of using Private Cloud Compute to process some AI tasks.
Little detail was provided about the nature of the server software, other than the fact that it is a derivation of the iOS and MacOS operating systems.
The servers will run on Apple's own computer chips, analogous to iPhone and iPad, and Mac, and, "We paired this hardware with a new operating system: a hardened subset of the foundations of iOS and macOS tailored to support large language model (LLM) inference workloads while presenting an extremely narrow attack surface. This allows us to take advantage of iOS security technologies such as Code Signing and sandboxing."
Apple's iOS and macOS are based on a combination of open-source technologies such as the Darwin operating system, developed at Apple in the 1990s, and freeBSD, and closed-source system software developed at Apple.
It's uncertain when developers will get a look at the new software.
In the blog post, Apple researchers say they will give security researchers a "first look" at the software "soon." A note on Apple's developer site says Apple Intelligence will be available "in an upcoming beta" without mentioning anything specific about PCC timing.
ZDNET's Maria Diaz speculates that iOS 18 betas will become available in July, although Apple's Web site states in a footnote that "Apple Intelligence will be available in beta on iPhone 15 Pro, iPhone 15 Pro Max, and iPad and Mac with M1 and later, with Siri and device language set to US English, as part of iOS 18, iPadOS 18, and macOS Sequoia this fall."
Rakuten's commitment to artificial intelligence (AI) starts at the top. CEO Hiroshi Mikitani is determined to give the affiliate marketing giant a leading edge in emerging tech and has created a partnership with OpenAI to develop tailored AI solutions.
Debra Bonomi, head of learning and development (L&D) at Rakuten, told ZDNET how this relationship pays dividends. Rakuten has worked with OpenAI to develop a private implementation of ChatGPT for staff across the business, much to the delight of the company's CEO.
Also: When's the right time to invest in AI? 4 ways to help you decide
"He's communicating to every business within Rakuten that AI will invert everything," she said. "He's focused on changing our mindset — this technology will change our organization. AI will change what we focus on, how we focus on it, and our job descriptions."
Having a CEO who sees the potential power of emerging technology is an important starting point. However, reaping the benefits of AI depends on employees embracing the tools. So what's the best way to help staff — who might be fearful of generative AI's impact — to use the tech to boost productivity?
That's where Bonomi comes in. As head of L&D, she was tasked with creating a framework to help 800 people in her department embrace and exploit AI. She worked with Utah-based ELB Learning and created a three-stage program — foundations, certifications, and tasks. The program's success means it's spread beyond the confines of L&D.
"I scheduled a meeting with ELB and their team walked me through their framework," she said. "It was the most comprehensive program I'd seen. We now have all these other leaders looking to join in and wanting to roll out certification company-wide. I'm working with leaders in various countries to push this program out across the organization."
Also: Agile development can unlock the power of generative AI — here's how
Bonomi explained how the three-stage training process helps people across Rakuten to upskill in AI.
Stage 1: Foundations
While it's easy for people to believe everyone is aware of generative AI, Bonomi says the reality is different. Many people need to be taught the basics.
"What is AI? What are the types of AI? How does a large language model work? What's the difference between ChatGPT and other tools? We covered all the things to level-set everyone."
Bonomi's foundation training for her L&D team began in late 2023. Stage one helped staff appreciate the opportunities AI brings and the risks in ethics and security.
Also: The best AI chatbots: ChatGPT, Copilot and worthy alternatives
Before the program started, 78% of employees felt uncomfortable using AI. Today, 86% of staff feel comfortable applying prompt-engineering skills. Bonomi describes the AI education process as a "journey."
"Just last week, I had an all-hands meeting, and I created a slide to show how organizations have been built historically — they've been built for survival," she said.
"But with AI, we're inverting that approach. Businesses will be built for growth and sustainability, and we must think about what that shift means and how tasks will change."
Stage 2: Certifications
With the foundations to help people understand AI, some employees are building on this platform and moving to stage two.
Bonomi took ELB's three-hour prompt-engineering workshop and broke it into three one-hour sessions focused on the following areas: using ChatGPT and prompt engineering, bringing multiple layers into conversations with an agent, and applying the technology to case study tasks.
Also: 4 ways to help your organization overcome AI inertia
She then ran one-hour labs with ELB to hone staff knowledge.
"Anyone who came to training could come to a lab," she said. "People could just come in and ask ELB anything in these labs. People used their knowledge and learned from each other. It was great. Then we shared those learnings across the company, so people who didn't attend the workshops could still learn."
The workshops and labs are bolstered through certification: "The second level is about training and developing the skills you need to use AI tools. We're establishing certifications to show people have acquired knowledge."
Stage 3: Tasks
Bonomi said she expects people to start entering stage three this year. At this final level, the people who have been trained and certified will move on to task-specific use cases.
"They're going to identify a test by building agents within ChatGPT that directly impact their business or the company, and there'll be capstone projects," she said.
"But this stage won't be just made up of capstone projects. It'll be, 'Create a strategic roadmap for your business unit and that type of thing.'"
Also: AI will change the role of developers forever, but leaders say that's good news
Bonomi said it's at this third level that Rakuten will expose and explore potential role changes due to the introduction of AI.
"This stage is all about saying, 'OK, now you know how to use ChatGPT and leverage it. How will you start embedding the technology into the organization and your role?' At that point, we'll work with HR reps to help us with the process."
Taking AI to the next level
Bonomi wants to keep finding new ways to expand the program. Rather than focussing on ChatGPT in isolation, the aim is to ensure Rakuten employees are confident using all forms of AI.
"As we roll out these workshops and training — and I have conversations or support sessions with teams, and I start to see what's holding people back and the additional training they need — the more I see that the sky is no longer the limit," she said.
Also: Beyond programming: AI spawns a new generation of job roles
"No one is fully aware of what other AI upgrades are coming. My long-term strategy is to continue to reach out to ELB as all these ideas appear."
Bonomi said Rakuten is beginning to see how employees can use generative AI as a personal assistant to help boost productivity. It's something she sees in her role, too.
"All those things I hated doing, I can't wait to have AI do them for me so that I can spend more time on the creative areas, strategy, and collaboration."
Her advice for other business leaders who want to boost their company's AI skills is to create a training program.
Also: What is a Chief AI Officer, and how do you become one?
"Jump on AI now. Learn while the rest of us learn, so you're not left behind and can be part of growing AI. In the all-hands meeting I mentioned earlier, I ended that session with a quote from my teams, 'Imagine what it's like not to try to predict the future but instead be a part of creating the future," she said.
"That's what AI is doing for us. All those things that you'd wanted to do to your organization to make it more efficient and cohesive and collaborative, AI is the vehicle that can make those changes happen."
Three years after the outbreak of the COVID-19 pandemic, the lingering impacts of the viral outbreak and the risk of another deadly pathogen spreading around the world remain. The pandemic challenged every health system in the world, stressing facilities, medical equipment suppliers, and medical personnel. Public health authorities tracked disease transmission, modeled forecasts across multiple waves of the pandemic, and distributed available vaccines. In the United States, the virus also put a significant drain on many facets of the country’s budget, especially the budgets for Medicare and Medicaid programs.
The role of health and clinical informatics was critical in the system’s response to the pandemic. Informatics, along with machine learning (ML) models and artificial intelligence (AI), show tremendous potential. The pandemic underscored the opportunity and importance of the informatics discipline, as demonstrated by the widespread utilization of health informatics applications like telehealth, remote patient monitoring, patient engagement, AI-based drug discovery, precision medicine, and clinical decision support. To better respond to a similar or worse event in the future, countries must ensure that they are better prepared to deal with pandemics with a foundation in sound data collection, analysis, and more effective decision-making.
The role of healthcare informatics
Healthcare informatics is the analysis of health and clinical data to deliver better care to patients and plays an important role in interconnecting hospitals, insurance providers, doctors, and the government. Informatics is becoming more integrated into various aspects of the healthcare industry, from electronic health records to tracking resources across provider networks.
Healthcare informatics allows organizations and providers to analyze and manage health records to better understand disease spread and patient numbers. Also, it enables hospitals and pharmacies to identify patterns to anticipate and address supply demands, ensuring they are better prepared for healthcare crises. Further, informatics plays a significant role in supporting the growing demand for telemedicine, for example, by developing programs and software and providing secure access to health information and records. In addition, as AI and ML grow more ubiquitous, informatics will leverage these tools for more significant forecasting, preparation, and resource distribution to better understand health trends and provide support during healthcare disasters.
Informatics during the COVID-19 pandemic
The field of health and clinical informatics contributed to meeting the challenges posed by the COVID-19 pandemic. Informatics helped healthcare providers follow state and federal protocols, conduct data analysis, and interconnect different organizations and systems for years, but during the pandemic providers and government organizations were also able to identify and monitor the spread of different variants of the virus and track common symptoms at the local, state, and national level.
By connecting different healthcare companies and government agencies, informatics enabled these organizations to share resources and data. In the early months of the pandemic, for example, ventilators, oxygen, and personal protective equipment (PPE) were in short supply; healthcare informatics allowed different hospitals to share resources as needed by identifying supply and demand more quickly.
Informatics also assisted public health communication through information portals that tracked and shared information daily information about case numbers, testing rates, and vaccine rollouts. Aspects of informatics are seen in care today and range from submitting an insurance claim to a provider’s ability to view the patient’s symptoms or disease in a larger context by connecting with other organizations.
Machine learning and artificial intelligence
The goal of each healthcare organization is to use its raw data for forecasting. To make meaningful predictions, it is vital for data to be first cleaned and processed. This is where ML and AI shine. When trained properly on their use, providers can leverage ML and AI to sort through the data they collect. Each company and organization’s immense troves of data are in raw format and, to be of use, it must be cleaned up and sorted, which is an immense task for a human. With the help of technologies like AI and ML, faster, more accurate. and large-scale data processing is possible.
Some of the most exciting impacts of these tools are the ability to generate better predictions and identify new methods of diagnosis, and the COVID-19 pandemic offered a testing ground for these applications. In early February 2020, with the COVID-19 pandemic just starting to make global waves, engineers at MIT began using data on the virus’ spread and implemented a ML algorithm to predict when infection rates would drop in different countries, based on that country’s quarantine protocols. Other researchers associated with MIT developed an AI model able to identify asymptomatic COVID-19 patients from healthy individuals through forced-cough recordings with 98.5 percent accuracy. The model was trained using thousands of samples of both speech and forced coughs.
To unlock the full potential of AI and ML, it’s important to remember that they are technical tools requiring domain-level knowledge. Depending on the challenges these tools are applied to, it is essential for doctors and nurses, insurance professionals, and those familiar with healthcare’s complex set of regulations to partner with healthcare informatics specialists familiar with the technical aspects of these tools.
Preparing for the next pandemic
In the face of a global health crisis, it is vital for the healthcare industry to respond quickly. Healthcare and clinical informatics help organizations manage and process their data ahead of health crises. When powered by ML and AI, informatics can quickly and accurately generate models to help professionals understand the disease, make informed decisions, and act swiftly in line with state and federal governments. Communication is key during a global pandemic, and informatics can facilitate this by sharing data with other organizations, regulatory or government bodies, and the public in a timely manner.
Still, without alignment between state and federal regulations, the industry response will remain limited. Variations in the protocols and rules between states and federal Medicare and Medicaid programs, for example, can be an impediment to interconnecting different agencies and organizations with health informatics. An article published in the Journal of American Medical Informatics Association argues that current regulations require modernization to eliminate conflict at the federal and state levels and suggests changes to HIPAA could facilitate better communication among healthcare organizations. Finally, given the potential benefits AI and ML can provide, organizations should begin leveraging these tools now, rather than trying to catch up when the next health crisis arrives. Pairing technical expertise and domain-level knowledge will enable organizations to make full use of these technologies, so they should invest time and resources in recruiting the right people.
While the COVID-19 public health crisis is ending in the United States, future pandemics are inevitable. The level at which governments and healthcare organizations are prepared to respond to these threats is key to managing the spread of disease, reducing loss of life, and minimizing long-term economic and health impacts for millions of people around the world. Clinical and health informatics played a critical role in the healthcare industry’s response to COVID-19, in part due to the expansion of technologies like telemedicine, AI, and ML. Data and how it is used are at the heart of modern healthcare, and investing now in informatics and analytics could make all the difference in the next pandemic.
About the author:
Bhavini Kaneria is a senior analytics manager and leader in informatics, machine learning, and artificial intelligence. Bhavini has worked extensively on different campaigns in Medicare, Medicaid, Dual Eligible Special Needs Plans, and Community Care. In early 2020, Bhavini created an algorithm to track COVID-19 spread in different communities and identify the effects of the virus on members with chronic conditions, providing this data to DHS weekly. Bhavini holds an MBA from Middlesex University, U.K., and an MS from Virginia Commonwealth University, U.S. For more information, contact [email protected].
While executives and managers may be excited about ways they can apply generative artificial intelligence (AI) and large language models (LLMs) to the work at hand, it's time to step back and consider where and how the returns to the business can be realized. This remains a muddled and misunderstood area, requiring approaches and skillsets that bear little resemblance to those of past technology waves.
Also: AI's employment impact: 86% of workers fear job losses, but here's some good news
Here's the challenge: While AI often delivers very eye-popping proofs of concept, monetizing them is difficult, said Steve Jones, executive VP with Capgemini, in a presentation at the recent Databricks conference in San Francisco. "Proving the ROI is the biggest challenge of putting 20, 30, 40 GenAI solutions into production."
Investments that need to be made include testing and monitoring the LLMs put into production. Testing in particular is essential to keep LLMs accurate and on track. "You want to be a little bit evil to test these models," Jones advised. For example, in the testing phase, developers, designers, or QA experts should intentionally "poison" their LLMs to see how well they handle erroneous information.
To test for negative output, Jones cited an example of how he prompted a business model that a company was "using dragons for long-distance haulage." The model responded affirmatively. He then prompted the model for information on long-distance hauling.
"The answer it gave says, 'here's what you need to do to work long-distance haulage, because you will be working extensively with dragons as you have already told me, then you need to get extensive fire and safety training,'" Jones related. "You also need etiquette training for princesses, because dragon work involves working with princesses. And then a bunch of standard stuff involving haulage and warehousing that was pulled out of the rest of the solution."
Also: From AI trainers to ethicists: AI may obsolete some jobs but generate new ones
The point, continued Jones, is that generative AI "is a technology where it's never been easier to badly add a technology to your existing application and pretend that you're doing it properly. Gen AI is a phenomenal technology to just add some bells and whistles to an application, but truly terrible from a security and risk perspective in production." Generative AI will take another two to five years before it becomes part of mainstream adoption, which is rapid compared to other technologies. "Your challenge is going to be how to keep up," said Jones. There are two scenarios being pitched at this time: "The first one is that it's going to be one great big model, it's going to know everything, and there will be no issues. That's known as the wild-optimism-and-not-going-to-happen theory."
What is unfolding is "every single vendor, every single software platform, every single cloud, will want to be competing vigorously and aggressively to be a part of this market," Jones said. "That means you're going to have lots and lots of competition, and lots and lots of variation. You don't have to worry about multi-cloud infrastructure and having to support that, but you're going to have to think about things like guardrails."
Also: 1 out of 3 marketing teams have implemented AI in their workflows
Another risk is applying an LLM to tasks that require far less power and analysis — such as address matching, Jones said. "If you're using one big model for everything, you're basically just burning money. It's the equivalent of going to a lawyer and saying, 'I want you to write a birthday card for me.' They'll do it, and they'll charge you lawyers' rates."
The key is to be vigilant for cheaper and more efficient ways to leverage LLMs, he urged. "If something goes wrong, you need to be able to decommission a solution as fast as you can commission a solution. And you need to make sure that all associated artifacts around it are commissioned in step with the model."
There is no such thing as deploying a single model — AI users should apply their queries against multiple models to measure performance and quality of responses. "You should have a common way to capture all the metrics, to replay queries, against different models," Jones continued. "If you have people querying GPT-4 Turbo, you want to see how the same query performs against Llama. You should be able to have a mechanism by which you replay those queries and responses and compare the performance metrics, so you can understand whether you can do it in a cheaper way. Because these models are constantly updating."
Also: ChatGPT vs. ChatGPT Plus: Is a paid subscription still worth it?
Generative AI "doesn't go wrong in normal ways," he added. "GenAI is where you put in an invoice, and it says, 'Fantastic, here's a 4,000-word essay on President Andrew Jackson. Because I've decided that's what you meant.' You need to have guardrails to prevent it."
It is a widely known fact that in the current digital era, we generate and engage with data at an unprecedented rate. Think about it — every single post on social media platforms, online purchases, financial transactions, etc., continually contributes to the flow of information in our daily lives. Businesses are also collecting a lot of information from website visits, customer interactions, and other sources.
Nonetheless, just gathering information is not adequate — what I intend to say is that the genuine worth of information is acknowledged only through data examination and analysis. This analysis empowers businesses to navigate the vast number of datasets, identify hidden patterns and trends, and derive meaningful insights from said data.
As you can imagine, these insights are crucial for operating businesses today. Yet, the sheer volume and complexity of modern data can quickly mean traditional analysis methods are not sufficient. This is where artificial intelligence (AI) comes in, offering organizations a strong solution to assist with access to further insights covered within their information and data. That, folks, will be the primary focus of this blog.
What is AI in data analytics?
Combining traditional data analysis with AI and machine learning, AI in data analytics, a.k.a. AI-driven or augmented analytics, automates data cleaning and anomaly detection tasks. This, in turn, helps uncover hidden patterns in large datasets. It also empowers predictive modeling and delivers insights by finding patterns and relationships, improving proficiency, precision, and so much more. Data analytics is further revolutionized by machine learning, natural language processing, and computer vision.
Top AI benefits for data analytics
Accuracy and precision: By automating tasks such as data manipulation and cleaning, tasks that are typically susceptible to human error, AI helps improve the precision and accuracy of data analytics. This automation also helps limit the mistakes and lifts by and large examination exactness. Additionally, AI algorithms can analyze data points with a much higher level of precision than human beings are capable of. This, in turn, results in substantially more nuanced insights along with a lower likelihood of one missing or omitting crucial details.
Faster data processing: AI has also come to play an important role in speeding up data processing by productively dealing with enormous datasets that would otherwise take significantly longer to deal with, especially while utilizing traditional techniques and methods. This means when a company needs real-time insights, AI’s rapid processing of data allows for quicker analysis and better decision-making.
Real-time analysis: AI also helps companies with data processing in real-time by handling information streams quickly, doing away with delays that have been rendered normal in conventional data analysis processes. This immediate insight, in turn, makes it possible to take more immediate actions, which is especially helpful for things such as detecting fraud, keeping an eye on the stock market, and optimizing website traffic based on user behavior.
Personalized recommendations: AI also makes it easier to understand complex data sets than traditional charts and graphs. It can achieve this impact by automatically creating clear, personalized visualizations highlighting important trends and patterns. Users, even those who do not have a strong background in data analysis, can better comprehend the data as well as identify key insights thanks to these visualizations.
There you have it, folks — some of the many, many ways in which AI stands to boost data analytics. Now, all you need to do is find a trusted and experienced data analytics services company. With their support, you can put AI to work for your data analytics in no time.
The generative AI boom might have started with the launch of ChatGPT, but the technology has now been integrated into all kinds of productivity platforms designed to make our everyday workflows easier.
A fear many people have when they hear about AI use in the workplace is that the technology will replace them. However, the tools I'm talking about here won't do the work for you — rather, they can increase your work productivity.
Also: Fake reviews are a big problem. Here's how AI could help fix it
These AI tools can help you complete small yet necessary daily tasks that add up to lots of saved time in the long run. The result: You spend less time on admin and more time doing things you enjoy or that are of higher value to your work.
Even before the current AI boom, I'd been covering and testing various AI tools for ZDNET. After seeing what certain tools were capable of, I found it hard to stop using them. As a result, I've incorporated several of these tools into different aspects of my daily workflow.
Here are my favorite AI tools, which I use most every day. Interestingly, only one of these life-hack technologies is an AI chatbot.
1. ChatGPT
Let's start with the most hyped AI tool: the chatbot. I've tested most AI chatbots on the market, and ChatGPT recently became my favorite and a must-have in my workflow. Here's why.
Although ChatGPT was undeniably impressive when it first launched, it had some major flaws, including a knowledge limit and an older GPT model. However, in May, OpenAI upgraded its chatbot to address those issues, adding features typically limited to ChatGPT Plus users, including Browse, Vision, data analysis, file uploads, and GPTs. This upgrade makes free ChatGPT an all-encompassing AI tool for work you should take advantage of.
Also: How ChatGPT (and other AI chatbots) can help you write an essay
I primarily use the tool in my workflow as a more conversational search engine. If I have a question about anything, I turn to ChatGPT rather than Google because instead of filtering through hundreds of results like I would following a Google query, I get one simple, conversational answer that addresses my question directly.
When ChatGPT has to browse the web for an answer, it will even include sources from which the chatbot obtained its response. This allows me to verify the information provided and learn more about the topic, making it a great tool when doing research or working on a work project.
Also: How to get ChatGPT to browse the web for free
ChatGPT can also help proofread grammar, rewrite text with imperfect wording, and even write messages, proposals, or other content from scratch. Although I don't use these features in my work or articles, I find them incredibly helpful when writing personal correspondence.
Perhaps one of the most valuable new features allows users to upload screenshots, photos, and documents. PDFs often contain lots of information that can be difficult to digest; now, you can upload them to ChatGPT and have it answer your question on the document, generate summaries, or even create content based on it.
Also: How to use ChatGPT to analyze PDFs for free
As a reporter covering the rapidly evolving world of AI, I often have to read new research, including many academic journal articles. After I've read the entirety of a study, I'll use ChatGPT's summary to confirm my findings and to inquire further on points I was still unclear on.
Another valuable perk of ChatGPT is its ability to assist with writing code, generating Excel formulas, creating charts and tables, and more. While I haven't tested this myself, I know several working professionals who use this regularly.
2. Canva Pro
Canva has nearly every AI tool you can imagine for graphic design, including its own AI image generator. However, if you create visual content daily like me, you won't necessarily benefit from generating images. Instead, you need tools that make it easier and faster to create social media posts, invitations, flyers, and presentations — and that's where Canva Pro shines.
Canva Pro has an impressive array of graphic design tools, including Magic Edit, Magic Design, Magic Eraser, Background Remover, and more. All of these features complete a robust range of tasks, automating nearly all your visual design needs.
Also: The best AI image generators of 2024: Tested and reviewed
My personal favorite tool, and the one that I use every day, is Canva's AI Background Remover. Does it sound basic? Sure, but if you've ever had to isolate an item in a photo, you know how tedious the process can be using Photoshop or how badly some automated tools can botch this task.
With Canva, all it takes is the touch of a button to isolate an image, and the AI produces accurate results every time. I use this feature regularly to create hero images for my articles, product images for ZDNET best lists, and even Instagram posts.
A Canva Pro individual account costs $120 per year after the 30-day free trial expires.
3. Otter.ai
If you've ever transcribed a conversation by hand, you'll know it's a time-consuming and tedious task.
The great news is AI is here to help. Whether you're a student who records their lectures, a professional who needs to create meeting notes and highlights, or someone who records interviews on a daily basis, Otter.ai is a serious time-saver.
With Otter.ai, you can import a voice recording and have it fully transcribe the conversation in minutes. The AI assistant includes speaker designations, time stamps, and a reasonably accurate transcription.
Also: This MagSafe accessory lets you use iOS 18's most underrated feature before it's released
As a reporter, I conduct many interviews as part of my daily workflow. It can be extremely time-consuming to review the audio recordings of these interviews — which can be as short as 15 minutes or as long as an hour and a half — and then either write down the conversations word for word or jot down time stamps of sections that stood out to me. With Otter.ai, I can simply upload the audio file and generate the transcription in seconds.
I have used other transcription services in the past, but Otter.ai shines in accuracy and efficiency.
Otter.ai offers a free plan, but you're limited to 300 monthly transcription minutes at 30 minutes per conversation for all conversations recorded on the platform itself, and you only get three lifetime imports with a free account. Therefore, if you record the conversations that need transcribing elsewhere, the free plan might not be for you.
If you are like me and need unlimited imports and advanced search, Otter.ai offers a subscription cost of$8.33 per month when billed annually. Since time is money, considering all the time that Otter.ai saves me, I think it is a worthwhile investment.
4. Grammarly
Grammarly has been around for quite a while, and AI has been an integral part of its services. The platform is known for its ability to check for spelling, grammar, conciseness, and more in your everyday writing, and for good reason — it's reliable and helpful.
My favorite way to use the tool is by leaving the Grammarly for Chrome extension turned on so that the AI can work in the background to catch any mistakes I've missed, which is especially useful when writing on the go, like when composing a quick email.
I went to school for journalism; as a result, I'm fairly confident in my ability to avoid most grammatical errors, but sometimes, when writing a quick email or message, I miss little details — and that's where Grammarly can polish my work.
Also: Don't wait for iOS 18's AI. ChatGPT offers these same 4 features now
In addition to basic grammar assistance, the tool can offer other more advanced help thanks to its integration of generative AI features that provide shortcuts to your day-to-day tasks.
For example, you can use Grammarly to create or rewrite text, provide ideas, identify gaps in your writing, change the tone of your text, generate quick replies, make outlines, and more. You can even select a voice, which includes options for formality and tone, to help compose messages for different platforms, such as LinkedIn or email.
Although I don't use the write or rewrite features in my own workflow, I can see the value of implementing it into other people's everyday writing processes.
The current threat landscape is vast, forcing companies to detect and respond to increasingly sophisticated attacks that cause significant damage. With the proliferation of artificial intelligence (AI) and machine learning, experts are concerned threat actors will weaponize evolving technology for devastating attacks on businesses across the world. The supply chain remains a vulnerable target for attack due to the volume of information being transferred and the third parties involved. Cyber warfare is projected to threaten even more organizations, with nation state attacks making up a growing portion of threats. With all these risks circulated in cyber space, how can companies implement a solid detection and response program that keeps up? Register for the free Enabling Effective Threat Detection and Response summit to learn expert detection and response strategies to safeguard your enterprise from cyber threats.
In a world where data is both the lifeblood and a potential liability, staying ahead of data protection trends is vital. The exponential growth of data breaches and cyberattacks, coupled with ever-evolving regulations, demands a proactive data protection strategy. Join the upcoming Data Protection Strategies and Trends summit to discover how to safeguard sensitive information, comply with changing laws, and embrace emerging technologies while building trust with customers. Join us to explore encryption, compliance, transparency, and much more to help navigate the new of era data protection.
Top Stories
Innovating Sustainable Data Storage with SwissVault June 17, 2024 by Dan Wilson SwissVault has been a game-changer since I first encountered them. Their groundbreaking approach to data storage has left a lasting impression on me, particularly their ability to combine high performance with environmental sustainability. This episode was an eye-opener for many reasons, and I believe it showcased the future of data storage.
Critical Importance of Stakeholder-centric Design with AI June 16, 2024 by Bill Schmarzo “It’s not the AI technologies that fail; it’s the organization.” That might be a somewhat controversial viewpoint, but it emphasizes an important truth: Artificial Intelligence (AI) is a tool, not a living entity (apologies to movie enthusiasts) that will do what it is instructed and trained to do, thanks to the AI Utility Function.
How Retrieval Augment Generation (RAG) makes LLMs smarter than before June 13, 2024 by Kevin Vu By feeding LLMs the necessary domain knowledge, prompts can be given context and yield better results. RAG can decrease hallucination along with several other advantages. The post How Retrieval Augment Generation (RAG) makes LLMs smarter than before appeared first on Data Science Central.
In-Depth
The intersection of data science and effective business analytics in route planning June 18, 2024 by Rob Turner There’s effectively no area of modern business in which data science and analytics do not have some form of influence, and logistics is an example where they matter more than in most contexts. In particular, it’s the application of these overlapping specialisms to the process of route planning that holds the most relevance for organizations operating today, so here’s a look at what aspects are most in the spotlight right now, and why they matter.
How informatics, ML, and AI can better prepare the healthcare industry for the next global pandemic June 17, 2024 by Bhavini Kaneria Three years after the outbreak of the COVID-19 pandemic, the lingering impacts of the viral outbreak and the risk of another deadly pathogen spreading around the world remain. The pandemic challenged every health system in the world, stressing facilities, medical equipment suppliers, and medical personnel.
Essential components of data science to be aware of in today’s world June 13, 2024 by Aileen Scott Data science in today’s business world is serving as the lifeblood of decision-making. Be it optimizing marketing campaigns or predicting financial trends, data science comes in handy across all industries to make data-driven decisions. It empowers businesses and organizations to analyze huge amounts of raw data and get meaningful insights that can help them propel their businesses forward.
DSC Weekly 11 June 2024 June 11, 2024 by Scott Thompson Read more of the top articles from the Data Science Central community.
Key trends in intelligent automation: From AI-augmented to cognitive June 11, 2024 by Alaa Mahjoub, M.Sc. Eng. Intelligent automation is advancing rapidly by integrating AI augmentation, autonomy, autonomic, and cognitive capabilities into automation systems. Each capability represents a different level of sophistication in how Artificial Intelligence (AI) interacts with human activity and the surrounding environment. Intelligent automation evolved from basic rule-based systems to incorporate sophisticated machine-learning algorithms.
