January 04, 2024 02:30 AM Eastern Standard Time

NEW YORK & AMSTERDAM–(BUSINESS WIRE)–With this latest innovation, users can now generate a complete audio ad simply by uploading a picture such as a product image, billboard ad, or even a photo of a storefront. This cutting-edge feature leverages the latest AI technology to analyze visual elements to create an engaging script before selecting suitable AI voices, music and sound effects to deliver a fully produced audio ad.

“Adthos is committed to revolutionizing the way audio advertising is produced”Post this

The platform uses AI to analyze the content of a picture, identifying brands, slogans, styles, target audience and much more to write a creative brief. From the creative brief an ad script is created, voices, music and sound effects are curated, before mixing all the elements together in a matter of minutes.

“Adthos is committed to revolutionizing the way audio advertising is produced,” says Raoul Wedel, CEO of Adthos. “Our new feature is a game-changer, instantly unlocking the potential of audio advertising for anyone that can take a picture“

Adthos Creative Studio’s new feature is an exciting addition to the Self-Service portal, designed to streamline the ad creation process and bring the power of AI to businesses of all sizes. Whether a seasoned marketer or a start-up business owner, anyone can leverage this feature to create dynamic and engaging audio ads that resonate with their target audience.

The makers of Adthos have created a short video introduction to the feature to provide more insight on the possibilities. Those interested in experiencing the creative power of this new feature for themselves can apply for a free trial via the website.

END

About Adthos:

Adthos is a leading AI Audio Platform, utilizing the latest in AI voice, text-to-speech and other AI technologies. The company is dedicated to developing innovative tools that help publishers and broadcasters and content creators streamline their processes and expand their reach to global audiences. For more information, visit www.adthos.com or contact press@adthos.com.

Contacts

press@adthos.com

https://www.businesswire.com/news/home/20240103724807/en/Adthos-Uses-AI-to-Create-Fully-Produced-Audio-Ads-From-a-Picture

The post Adthos Uses AI to Create Fully Produced Audio Ads From a Picture first appeared on AITechTrend.

Diffusion Bee: Simple Text-to-Image Generation for M1 Mac

Diffusion Bee is an incredible tool that allows you to run Stable Diffusion locally on your M1 Mac with just a single click. This hassle-free solution eliminates the need for complex setups and technical knowledge. By providing a one-click installer, Diffusion Bee ensures a seamless experience for generating images from your text prompts. With no data sent to the cloud except for weight downloads and software updates, your privacy is safeguarded.

Stable Diffusion UI: Browser-Based Image Generation

Stable Diffusion UI offers another convenient one-click installer, granting you access to a user-friendly browser interface for generating images from text and image prompts. Simply enter your text prompt, and witness the magic of image generation unfold before your eyes. While currently not compatible with Mac, Stable Diffusion UI runs seamlessly on Windows 10/11 and Linux.

Charl-E: Simplified Application for Image Generation

If you seek a straightforward and user-friendly solution for text-to-image generation, Charl-E is the answer. This application packages Stable Diffusion into a simple yet powerful tool. With Charl-E, you can effortlessly download the application and start exploring your creative vision without the need for complex setups, dependencies, or an internet connection.

NMKD Stable Diffusion GUI – AI Image Generator

For local hardware-based text-to-image generation, NMKD Stable Diffusion GUI is an excellent choice. With this ML toolkit, you can unleash your creativity on Nvidia GPUs. Please note that AMD GPUs are currently not supported. Ensure your system meets the minimum requirements, such as an Nvidia GPU with 4 GB VRAM and 8 GB RAM.

ImaginAIry: Pythonic Generation of SD Images

ImaginAIry offers a pythonic approach to text-to-image generation using Stable Diffusion. By installing it with a simple pip command, you gain access to a wide range of features, including memory efficiency improvements, prompt-based editing, face enhancement, upscaling, tiled images, and more. Compatible with Linux and macOS (M1), ImaginAIry is a versatile tool that enables you to bring your creative ideas to life.

Mage Space: Unfiltered SD for Text-to-Image Generation

Mage Space provides unfiltered access to Stable Diffusion for text-to-image generation. Its latest feature, Image2Image, allows you to combine an image of your choice with your text prompt, unlocking endless creative possibilities. Visit the Mage Space website to explore this remarkable tool and experience the freedom of unrestricted image generation.

Dreamlike.art: Your Dream Destination for Image Generation

Dreamlike.art is currently offering free access to its image generation services for a limited time. If you require additional credits, you can easily acquire them on the “Buy Credits” page without any charges. This platform aims to make your creative journey hassle-free and cost-effective.

FindAnything.App: Simplify Image Search and Access

Finding the perfect images for your projects can be challenging, especially when it comes to avoiding copyrighted content and excessive expenses. With FindAnything.App, your search for high-quality images becomes effortless. This browser extension enhances your Google image searches by presenting novel images alongside the standard options.

Major SD Forks: Enhance and Customize SD Projects

Several major SD forks allow you to customize and enhance Stable Diffusion projects without impacting the original repositories. These options offer flexibility and versatility, enabling you to incorporate updates and submit changes effortlessly.

– Automatic1111 – SD Web UI: Feature-Rich Browser Interface

Automatic1111 – SD Web UI provides a browser interface based on the Gradio library, offering various modes for text-to-image and image-to-image generation. It includes features such as outpainting, inpainting, prompt matrices, Stable Diffusion upscale, and more.

– InvokeAI: Slick WebGUI and Command-Line Interface

InvokeAI presents a sophisticated WebGUI and an interactive command-line script that combines text-to-image and image-to-image functionality. With its “dream bot” style interface, InvokeAI offers a unique user experience. Whether you are on Windows, Mac, or Linux, you can take advantage of its multiple features and enhancements.

– Waifu Diffusion: Specialized for Anime and Manga Art

Waifu Diffusion is a project that focuses on anime and manga art, training the Stable Diffusion model on over 56,000 images from Danbooru, a popular drawing site. If you’re an anime enthusiast, this tool will cater to your specific needs.

– Basujindal: Optimized VRAM Usage

Basujindal provides an optimized version of Stable Diffusion that reduces VRAM usage while sacrificing inference speed. By dividing the Stable Diffusion model into parts and transferring them to the GPU only when necessary, Basujindal offers improved efficiency.

Support and Enhance Text-to-Image Generation Tools

These powerful tools built on the Stable Diffusion model have opened up new horizons for text-to-image generation. Each tool offers unique features and capabilities, catering to diverse requirements. Whether you’re a professional artist or an enthusiast, these tools empower you to create captivating visual content effortlessly. Support these projects by exploring their respective repositories, contributing to their development, and spreading the word about their exceptional capabilities.

The post Democratizing Art: Unlocking the Power of Text with Image Generation Tools first appeared on AITechTrend.

Introduction

AI is a powerful tool that has already transformed many industries, including healthcare, finance, and transportation. However, as AI becomes more prevalent, there is a growing concern that it may be misused, intentionally or unintentionally. In this article, we will examine some of the most common misuses of AI and their potential consequences.

Misuse 1: Biased AI

One of the most significant concerns regarding AI is bias. AI systems are only as unbiased as the data they are trained on, and if that data is biased, the resulting AI will be biased as well. Biased AI can have serious consequences, such as discriminatory hiring practices or biased decision-making in the criminal justice system.

Misuse 2: Autonomous Weapons

The development of autonomous weapons is another area of concern. These weapons can make decisions and take actions without human intervention, which could lead to unintended consequences or war crimes. There is a growing movement to ban the development and use of autonomous weapons to prevent their misuse.

Misuse 3: Deepfakes

Deepfakes are another emerging concern with AI. These are videos or images that are manipulated to create a false impression. While they have some legitimate uses, such as in the film industry, they can also be used for fraud or disinformation campaigns. Deepfakes can be difficult to detect, and their misuse could have serious consequences for individuals or even entire countries.

Misuse 4: Privacy Violations

AI can also be misused to violate privacy. For example, facial recognition technology can be used to identify individuals without their consent, or to track their movements. This can have serious consequences for personal privacy and safety, as well as civil liberties.

Misuse 5: Job Losses

One of the potential consequences of AI is job losses, particularly in industries that are heavily reliant on manual labor or repetitive tasks. While AI can increase efficiency and productivity, it can also lead to the displacement of human workers, which could have economic and social consequences.

Misuse 6: Cyber Attacks

AI can also be used to launch cyber attacks. For example, AI-powered bots can be used to spread malware or phishing emails, or to launch distributed denial of service (DDoS) attacks. As AI becomes more advanced, the potential for cyber attacks using AI will only increase, which could have serious consequences for businesses and individuals.

Misuse 7: Misinformation

Finally, AI can be misused to spread misinformation or fake news. This can have serious consequences for public health and safety, as well as for democracy itself. Misinformation campaigns can be difficult to detect and can spread quickly, which makes them a powerful tool for those who seek to manipulate public opinion.

Conclusion

AI has enormous potential to transform our lives for the better, but we must also be vigilant about its potential misuse. By understanding the top misuses of AI and their potential consequences, we can work to prevent these misuses and ensure that AI is used in a responsible and ethical manner.

The post Beware the Misuse of AI: 7 Ways AI Could Do More Harm than Good first appeared on AITechTrend.

Introduction to Reinforcement Learning

Before diving into OpenAI Gym, it is essential to understand the basics of reinforcement learning. Reinforcement learning is a subfield of machine learning that involves training agents to make decisions based on rewards or penalties. The agent learns by interacting with an environment, receiving rewards or penalties for its actions. Over time, the agent learns to take actions that maximize its rewards, resulting in a more efficient and effective decision-making process.

Installing OpenAI Gym

To get started with OpenAI Gym, we need to install it on our system. OpenAI Gym can be installed using pip, a package manager for Python. The following command can be used to install OpenAI Gym:

pip install gym

Once installed, we can verify the installation by importing the gym module in Python.

import gym

Exploring Environments in OpenAI Gym

OpenAI Gym provides a wide range of environments for testing and benchmarking reinforcement learning algorithms. These environments simulate various scenarios, such as games, physics simulations, and robotics.

To explore the available environments in OpenAI Gym, we can use the following command:

import gym
print(gym.envs.registry.all())

This command will print a list of all the available environments in OpenAI Gym. Each environment has a unique identifier, which can be used to create an instance of that environment.

Creating an Environment in OpenAI Gym

To create an instance of an environment in OpenAI Gym, we can use the make method provided by the gym module. For example, to create an instance of the CartPole environment, we can use the following code:

import gym
env = gym.make('CartPole-v0')

The CartPole-v0 environment is a classic control problem, where the goal is to balance a pole on a cart. The environment provides four observations, representing the position and velocity of the cart and pole, and two actions, representing the force to apply to the cart (left or right).

Interacting with the Environment

Once we have created an instance of an environment, we can interact with it by taking actions and receiving observations and rewards. The env object provides several methods for interacting with the environment, such as reset, step, and render.

The reset method initializes the environment and returns the initial observation. The step method takes an action and returns the next observation, reward, and a boolean flag indicating if the episode is done. The render method displays the current state of the environment.

import gym
env = gym.make('CartPole-v0')
observation = env.reset()
for t in range(1000):
    env.render()
    action = env.action_space.sample()
    observation, reward, done, info = env.step(action)
    if done:
        print("Episode finished after {} timesteps".format(t+1))
        break
env.close()

In the above code, we first create an instance of the CartPole-v0 environment and reset it to get the initial observation. We then enter a loop, where we take a random action, receive the next observation, reward, and done flag, and render the environment. The loop continues until the episode is done.

Building a Reinforcement Learning Agent

Now that we have explored the basics of OpenAI Gym and how to interact with an environment, let’s build our first reinforcement learning agent. In this example, we will use the Q-learning algorithm to train an agent to play the FrozenLake environment.

The FrozenLake Environment

The FrozenLake environment is a gridworld game, where the goal is to navigate an agent from the start position to the goal position without falling into holes. The environment provides a 4×4 gridworld, with four actions (up, down, left, right) available at each grid cell. The agent receives a reward of +1 for reaching the goal and a reward of 0 for falling into a hole. The environment is considered solved if the agent can reach the goal with an average reward of 0.78 or higher over 100 episodes.

Q-Learning Algorithm

Q-learning is a model-free reinforcement learning algorithm that learns the optimal action-value function for a given environment. The action-value function represents the expected reward for taking a particular action in a particular state. The Q-learning algorithm updates the action-value function using the Bellman equation:

Q(s, a) = Q(s, a) + α(r + γ max Q(s’, a’) – Q(s, a))

where Q(s, a) is the action-value function for state s and action a, r is the reward received for taking action a in state s, s’ is the next state, a’ is the next action, α is the learning rate, and γ is the discount factor.

Implementing Q-Learning in OpenAI Gym

To implement Q-learning in OpenAI Gym, we first create an instance of the FrozenLake environment and define the Q-table. The Q-table is a dictionary that maps each state-action pair to its action-value function. We then enter a loop, where we take an action based on the current state and the Q-table, receive the next state and reward, update the Q-table using the Q-learning equation, and render the environment. The loop continues until the environment is solved or the maximum number of episodes is reached.

import gym
import numpy as np

# Create FrozenLake environment
env = gym.make('FrozenLake-v0')

# Define Q-table
Q = np.zeros([env.observation_space.n, env.action_space.n])

# Set hyperparameters
alpha = 0.8
gamma = 0.95
epsilon = 0.1
num_episodes = 2000

# Train agent
for episode in range(num_episodes):
    state = env.reset()
    done = False
    t = 0
    while not done:
        # Choose action using epsilon-greedy policy
        if np.random.uniform() < epsilon:
            action = env.action_space.sample()
        else:
            action = np.argmax(Q[state, :])
        # Take action and receive next state and reward
        next_state, reward, done, info = env.step(action)
        # Update Q-table
        Q[state, action] = Q[state, action] + alpha * (reward + gamma * np.max(Q[next_state, :]) - Q[state, action])
        state = next_state
        t += 1
    # Decay epsilon
    epsilon = 1.0 / (episode + 1)
    # Print episode information
    if episode % 100 == 0:
        print("Episode {}: Steps = {}, Reward = {}".format(episode, t, reward))

# Test agent
total_reward = 0
for i in range(100):
    state = env.reset()
    done = False
    while not done:
        action = np.argmax(Q[state, :])
        state, reward, done, info = env.step(action)
        total_reward += reward
env.render()
if done:
print("Episode {}: Reward = {}".format(i, total_reward))
total_reward = 0

In conclusion, OpenAI Gym provides an easy-to-use platform for developing and testing reinforcement learning algorithms. By providing a standardized set of environments and interfaces, it allows researchers and developers to focus on creating intelligent agents without worrying about the underlying mechanics of the environment. In this article, we covered the basics of OpenAI Gym, including how to interact with environments, and we implemented a simple Q-learning algorithm to train an agent in the FrozenLake environment. With the knowledge and skills gained from this article, you can now begin exploring more complex environments and algorithms, and develop intelligent agents for a wide range of applications. Whether you are a seasoned machine learning expert or just getting started, OpenAI Gym provides a powerful tool for developing and testing intelligent agents, and is sure to play an important role in the future of AI research and development.

The post OpenAI Gym: A Powerful Tool for Developing and Testing Reinforcement Learning Algorithms first appeared on AITechTrend.

Introduction to AI Accelerators

Hardware AI accelerators are specialized processors that help speed up machine learning workloads, offering high performance and low power consumption. There are various types of hardware accelerators in the market, including GPUs, FPGAs, ASICs, and TPUs, each designed for specific workloads.

GPUs (Graphics Processing Units)

Graphics processing units (GPUs) are the most commonly used hardware accelerators for machine learning workloads. These processors have been in use since the late 1990s and are widely available and easy to program. GPUs offer high parallelism, making them ideal for deep learning workloads. Their architecture is well-suited to handle massive amounts of data simultaneously, making them ideal for high-performance computing (HPC) workloads.

FPGAs (Field Programmable Gate Arrays)

Field Programmable Gate Arrays (FPGAs) are programmable logic devices that can be configured to perform specific tasks, making them suitable for custom logic applications. Unlike GPUs, which are general-purpose processors, FPGAs can be designed to meet specific requirements. They offer low latency and high bandwidth, making them ideal for real-time processing applications like computer vision and speech recognition.

ASICs (Application-Specific Integrated Circuits)

Application-specific integrated circuits (ASICs) are custom-designed processors optimized for specific workloads. Unlike FPGAs, which can be reprogrammed, ASICs are designed for a specific purpose and cannot be reconfigured. ASICs offer high performance and low power consumption, making them ideal for AI workloads that require high throughput.

TPUs (Tensor Processing Units)

Tensor Processing Units (TPUs) are Google’s custom-designed AI accelerators optimized for deep learning workloads. TPUs are specifically designed to accelerate TensorFlow, Google’s open-source machine learning framework, making them ideal for applications like image and speech recognition. TPUs offer high throughput and low power consumption, making them an ideal choice for large-scale AI workloads.

Comparison of Hardware AI Accelerators

When choosing a hardware AI accelerator, it is essential to consider the workload requirements, power consumption, and performance. GPUs are the most widely used AI accelerators and offer high performance and low power consumption. FPGAs offer low latency and high bandwidth, making them ideal for real-time processing applications. ASICs offer high performance and low power consumption, making them ideal for specific AI workloads. TPUs are Google’s custom-designed AI accelerators and offer high throughput and low power consumption, making them ideal for large-scale AI workloads.

Advantages and Disadvantages of Hardware AI Accelerators

Hardware AI accelerators offer several advantages over traditional CPUs, including high performance, low power consumption, and lower latency. However, there are some disadvantages to using hardware AI accelerators, including higher costs, limited scalability, and the need for specialized skills to program them.

Conclusion

Hardware AI accelerators are essential for running AI workloads and offer several advantages over traditional CPUs. Choosing the right accelerator depends on the workload requirements, power consumption, and performance. GPUs are the most widely used AI accelerators and offer high

The post Deep Learning Accelerators: How to Choose the Right Hardware for Your Needs first appeared on AITechTrend.

Parascript, which has been delivering high-performing automation for over 25 years and processing more than 100 billion documents annually, released today a new version of CheckXpert.AI that again represents the latest advancements in Deep Learning applied to Payment Processing.

“If you’re looking for lights-out automation for check processing at the highest, and I mean literally the highest level of performance that can be achieved, then this new release of CheckXpert.AI is it,” said Greg Council, VP of Marketing & Product Management at Parascript, “with a straight-through processing rate of greater than 98% at 99% accuracy, there really isn’t a need to employ any manual effort anymore.”

With the new release of CheckXpert.AI, Parascript continues leveraging its proprietary deep learning algorithms. CheckXpert.AI processes checks in a significantly smarter, more human-like way. CheckXpert.AI takes care of the full stream of documents for Proof of Deposit (POD) and Remittance applications. Here are a few highlights:

Significant error rate reduction, therefore, accuracy increase, for amount recognition on checks.

Parascript CheckXpert.AI significantly increases the level of accuracy and reliability of output which translates to up to not only an extremely high rate of touchless automation, but a 75% further reduction in handling exceptions.

This means that CheckXpert.AI successfully processes 99% of checks with an accuracy that goes beyond the accuracy of that of human operators, and in some cases, outperforms the accuracy of double keying in some applications.

Improvements in reporting types of documents

The new version of Parascript CheckXpert.AI is able to more accurately detect the document type for the stream of documents being processed by banks – personal checks, business checks, deposit slips, cash tickets, money orders, etc. This allows for a more accurate document classification.

Improvements in location

Parascript CheckXpert.AI has been improved to perform better when locating payee and date fields on business checks. This allows for a more accurate payee reading results and reduction in keying.

The post Parascript Is Upping the Game in Intelligent Document Processing with a New Version of CheckXpert.AI® first appeared on AITechTrend.

Neo4j^®, the world’s leading graph data platform, crossed $100 million in annual recurring revenue (ARR) during 2021. The year was marked by strategic product innovation that drove customer and partner excellence, strong community engagement, and super-sized venture funding investments.

“Neo4j has pioneered the graph space for a number of years, with critical deployments among major credit card firms for fraud detection, as well as use cases in areas driven by the pandemic, including product testing and supply chain analysis,” said Carl Olofson, Research Vice President at IDC.

Neo4j continued to grow in popularity throughout 2021 as the world’s most widely deployed graph database, maintaining its position as a top 20 database overall. Momentum drivers include the accelerated adoption of Neo4j AuraDB, a fully managed service that reduces friction as complex applications shift to the cloud, as well as the success of Neo4j Graph Data Science, a complete toolset for data scientists to apply graph algorithms for more effective machine learning and better predictions.

Over 1,000 organizations depend on Neo4j for mission-critical applications, and many thousands more experiment, prototype, and deploy Neo4j’s expanding portfolio of cloud services. Notable customers include Pfizer, PepsiCo, Inc., World Health Organization (WHO), Cable News Network, Inc. (CNN), and BMW Group.

Neo4j’s success in helping customers across industries such as Financial Services, Retail, and Healthcare caught the attention of investors, leading to $390 million in new investments raised in 2021, and launching Neo4j to a $2 billion valuation. On top of being the largest single funding round to date in the database space, Neo4j also welcomed GV (formerly Google Ventures) as a strategic investor and added former Google CFO, Patrick Pichette, to its board to offer increased industry expertise for the next phase of growth.

Patrick Pichette, Inovia Capital Partner and Neo4j Board Member, touched upon Neo4j’s momentum over the past year.

“2021 marked an incredible year for Neo4j and graph technology at large,” said Pichette. “What really sets Neo4j’s graph technology apart is that it uniquely solves some of the world’s most complex challenges. Neo4j is poised for strong, consistent growth leading into 2022, and we’re excited to be part of that journey.”

Emil Eifrem, CEO and Co-Founder of Neo4j, reflected on the past year and leading one of only a handful of private database companies to cross $100 million in ARR.

“In 2021, we demonstrated that Neo4j is a mainstay of modern data infrastructure, grounded in a global community of developers and data scientists, empowered with a rich portfolio of technology to address complex challenges, and scale without barriers,” said Eifrem. “We enter 2022 with the wind at our backs, and the right talent and leadership in place. We’re poised to deliver Neo4j to a fast-growing user base, and continue to delight our customers as their use cases become more exacting.”

The company ended 2021 with over 600 employees, representing the largest collective of graph expertise in the world. During the course of the year, Neo4j expanded rapidly in Asia-Pacific (Shanghai, Singapore, Sydney, Jakarta, and Bangalore), and Latin America (São Paulo).

Notable Neo4j 2021 milestones include:

Technology Leadership

• Breaking the Graph Scale Barrier: As part of NODES 2021, Neo4j demonstrated its super-scaling technology to show real-time query performance against a graph with over 200 billion nodes and more than a trillion relationships, running on over one thousand machines.
• Graphs and AI: Neo4j Graph Data Science was adopted by over 50 customers to build sophisticated AI, machine learning, and advanced analytics applications.
• AuraDB Enterprise: The most deployed and trusted graph technology platform was made generally available as a fully managed service, helping organizations including Levi Strauss & Co. and Adeo to radically accelerate time to value and get to production faster.
• Knowledge Graphs Accelerate Adoption: Two-thirds of Neo4j customers – including NASA – are implementing knowledge graphs to redefine what’s possible in data management and analytics.

Demonstrable Customer Value

• Unsurpassed ROI: The Neo4j Graph Data Platform pays for itself more than 4x in the span of three years (417% ROI), according to a recent Forrester TEI report.
• Accelerated Time to Value: According to Forrester, Neo4j showed 60% accelerated time to value, as average development time shrunk from 12 months to four.
• Digital Transformation: The TEI study was based on Forrester’s in-depth interviews with Neo4j customers who realized substantial cost savings from IT modernization and rationalization.

Commercial Impact

• Neo4j on Azure, GCP, and AWS: Neo4j is now globally available on Microsoft Azure, Google Cloud Platform (GCP), and Amazon Web Services (AWS) marketplaces. Customers can now seamlessly deploy Neo4j on the cloud platform of their choice.
• New Executives and Board Members: Neo4j welcomed Kristin Thornby as Chief People Officer. Nathalie Kornhoff-Bruls of Eurazeo and Patrick Pichette of Inovia Capital both joined Neo4j’s board.
• Partner Traction: Neo4j trained and certified over 1,000 graph practitioners from leading global system integrators including Accenture, Deloitte, EY, Capgemini, and PwC, in addition to closing new business with nine U.S. Federal Programs. The company expanded its partner leadership in emerging markets including Brazil, China, India, and Australia.

Market Expansion

• Top 20 Database: Neo4j is the leading graph database and ranks among the 20 most popular databases globally, as tracked by DB-Engines, an independent ranking of all database management systems.
• Award Wins: Neo4j was named one of CRN’s Big Data 100 Coolest Database System Companies, 2021 Data Breakthrough Awards Graph DB Solution of the Year, and 2021 Datanami Readers’ and Editors’ Choice Awards for Best Data and AI Technology.
• Rise in Cybersecurity Graph Adoption: Companies like Deloitte and Uplevel Security increasingly sought to leverage graph databases to strengthen cybersecurity strategies in response to increasingly sophisticated threats. Graphs equip cybersecurity teams with the competitive advantage of rapid screening, improved turnaround time, quality of recommendations, and reduced false positives, ultimately allowing them to mitigate cyber threats.
• Inclusion in U.S. Department of Defense (DoD) Iron Bank: Neo4j collaborated with The DoD’s Platform One organization to develop a hardened container for its graph database technology and joins a repository of digitally signed, binary container images that have been hardened according to the Iron Bank’s Container Hardening Guide.

Community Engagement

• Growing Developer Base: The global Neo4j community surpassed 240,000 members over the last year. During 2021, developers downloaded Neo4j more than 36 million times and launched more than 150,000 Neo4j Sandbox instances. Upwards of 53,000 professionals list Neo4j as a skill on their LinkedIn profiles.
• The Pandora Papers: The International Consortium of Investigative Journalists (ICIJ) released the Pandora Papers, which used Neo4j to generate visualizations and make searchable records of the hidden riches of world leaders. Neo4j has been working with the ICIJ since the 2016 Panama Papers investigation.
• Graphs4Good: The efforts of the Neo4j community to collaborate and help fight against the spread of COVID-19 were recognized by two honorable mentions in the AI and Data and Software categories of Fast Company’s 2021 World Changing Ideas Awards.
• Largest Graph Event: Neo4j Online Developer Expo and Summit (NODES 2021) welcomed over 12,000 registrants to listen to presentations from Fujitsu Research Labs, Dataiku, BASF, Apiax, Linkurious, and more.
• 2021 Graphie Award Winners: This year’s nominations eclipsed all prior years, with Neo4j receiving nominations spanning more than 10 countries and awarding 27 winners including Pfizer, Qualicorp S.A., Commonwealth Bank of Australia, Lenovo, Volvo Cars, Levi Strauss & Co., and many more.

 Resources

• Neo4j ROI
• Neo4j Products
• Neo4j Customers
• Neo4j Community
• Neo4j on Twitter
• Neo4j on LinkedIn
• Neo4j on YouTube
• Neo4j is Hiring

About Neo4j
Neo4j is the world’s leading graph data platform. We help organizations – including Comcast, ICIJ, NASA, UBS, and Volvo Cars – capture the rich context of the real world that exists in their data to solve challenges of any size and scale. Our customers transform their industries by curbing financial fraud and cyber crime, optimizing global networks, accelerating breakthrough research, and providing better recommendations. Neo4j delivers real-time transaction processing, advanced AI/ML, intuitive data visualization and more. Find out more at neo4j.com and follow us at @Neo4j.

The post Neo4j Closes Banner Year Marked by Customer Successes, Continued Industry Validation, Community Engagement, and Major Funding first appeared on AITechTrend.

ASAPP is a research-based artificial intelligence software provider that solves large, complex, data-rich problems with AI Native® technology. Large enterprises use ASAPP to make customer experience teams highly productive and effective by augmenting human activity and automating the world’s workflows.

As Chief Scientist, I’m responsible for setting the direction of the research and data science groups in order to achieve ASAPP’s vision to augment human activity positively through the advancement of AI. The group is currently focused on advancing the field of task-oriented dialog in real-world situations like customer care. Our research group consists of machine learning and language technology leaders, many of whom publish multiple times a year. We also have some of the best advisors in the industry from universities like Cornell and MIT.

Can you tell us about your journey into this market?

I have spent the past 20 years working in natural language processing and machine learning. My first project involved automatically summarizing news for mobile phones. The system was sophisticated for its time, but it amounted to a number of brittle heuristics and rules. Fast forward two decades and techniques in natural language processing and machine learning have become so powerful that we use them every day—often without realizing it.

After finishing my studies, I spent the bulk of these 20 years at Google Research. I was amazed at how machine learning went from a promising tool to one that dominates almost every consumer service. At first, progress was slow. A classifier here or there in some peripheral system. Then, progress came faster, machine learning became a first-class citizen. Finally, end-to-end learning started to replace whole ecosystems that a mere 10 years before were largely based on graphs, simple statistics, and rules-based systems.

After working almost exclusively on consumer-facing technologies. I started shifting my interests towards the enterprise. There were so many interesting challenges that arose in this space. The complexity of needs, the heterogeneity of data, and often the lack of clean, large-scale training sets are critical to machine learning and natural language processing. However, there were properties that made enterprise tractable. While the complexity of tasks was high, the set of tasks any specific enterprise engaged in was finite and manageable. The users of enterprise technology are often domain experts and can be trained. Most importantly, these consumers of enterprise technology were excited to interact with artificial intelligence in new ways— if it could deliver on its promise to improve the quality and efficiency of its efforts.

This led me to ASAPP. The work we’re doing in augmenting the customer service agent experience and performance is immensely rewarding. How can AI improve the agent experience leading to less burnout, lower turnover, and higher job satisfaction? This is in an industry that employs three million people in the United States alone but suffers from an average of 40 percent attrition—one of the highest rates of any industry.

How AI is elevating human performance?

Our central hypothesis at ASAPP is that AI should not replace humans, but augment them in positive and productive ways. This vision is broad and we have ambitions to apply it to all relevant human activity. However, as this is a broad mandate, the first area we’ve chosen to focus on is the customer experience domain.

The customer experience domain embodies all the challenges and rewards that come with augmenting human activity. Agents are engaged in complicated problem-solving tasks that require them to follow workflows, retrieve relevant information from customer and knowledge bases, and adapt to nuanced situations that a customer might find themselves in.

This gives rise to a huge number of opportunities for AI to improve that process. However, we think it is important to do this in a positive way, by which we mean:

• Augmentation happens at points that are natural and fluid during the course of the agent’s job. This is critical. If AI is interfering or interjecting at awkward moments or with poor latency, this will actually have a negative effect on the agent’s experience as they will need to consciously ignore the AI.
• More critically, we want the AI to achieve positive outcomes for all humans involved. In this case it is the customer, the agent and the organization. Customers want their issues handled efficiently and effectively. Agents want to do that for customers. Additionally, agents are doing a hard job, often dealing with difficult unsatisfied customers. AI should help them balance work and cognitive load in order to decrease fatigue and burnout and increase job satisfaction. Afterall, agents at call centers have one of the worst attrition rates (as high as 100% annually in some call centers) of any job in America. Finally, we want positive business outcomes for the company who runs the call center. This can be customer satisfaction, the throughput of issues that can be handled in a day or even the amount of sales.

For call centers, we often think of the positive outcomes between the customer, agent, and company as being in conflict with each other. But good AI will help to optimize these outcomes for all three.

How AI is being used to help train new workers?

The emergence of AI technologies to augment their performance during a call or digital customer interaction is becoming more commonplace, but AI to train workers is presently less conventional. Today, many agents train on new issues or procedures ‘live’. That is, they get a description of the procedure, but then only see it in practice when they take a call with a real customer. Imagine we gave pilots the manual of the plane and then told them to fly 300 passengers to Denver? Because of this, we are focusing on using AI to help build tools for agents to practice procedures and handle difficult situations before they deal with live customers. When this is coupled with targeted feedback (either by a supervisor or automatically) this will allow the agent to grow their skills in a less stressful environment.

How ASAPP is using AI to reduce turnover and augment this new generation of workers?

Large companies offering consumer goods and services spend millions, and sometimes billions of dollars each year on contact centers that serve their customers, with the labor cost representing 80-90% of total costs. It’s a big problem driving agent turnover to be 40%—and sometimes 100% or more—every year.

There is often a misconception that agents are indifferent to your problems and are going through the motions. In the worst case, even obstructs your ability to solve a problem. Nothing can be further from the truth. Agents, as with all people, derive satisfaction from helping customers solve their problems. How would you rather spend your day, hearing robust ‘thank you’s or screaming customers? In a recent study we conducted, we found that 90% of agents reported that calls with customers made their day, and the majority say they are happy with their jobs. But, agents want the tools and training required in order to make customers happy. Unhappy customers lead to frustrated, fatigued, and stressed agents. This is the primary driver of turnover.

AI to augment the agents during a call already helps. If the agent has the tools and guidance on how to effectively and quickly solve a problem for a customer, then the odds that the customer is happy can only be higher, which in turn should lead to higher job satisfaction.

Better AI to improve customer satisfaction in dynamic situations as well as AI for grounded training — that is how ASAPP puts focus on the agent with the ultimate goal of reducing turnover.

After GPT-4, what will the future of large language models look like for the enterprise?

Every time we reach a new peak in what GPU/TPUs are able to handle, we see leading technology companies put out new, larger, pre-trained language models. These large, pre-trained models can be foundational for a number of downstream tasks and applications in NLP. While we’ll continue to use these pre-trained language models for the foreseeable future, one always needs to consider the value-added of a more powerful model vs. the costs that come with training and deploying it, especially in the enterprise.

For enterprise uses and applications, the future of innovation is now centered around the fine-tuned and specialized models created from these large language models to be the world’s best for a specific application or domain. It’s great to see how Hugging Face has democratized access to these large language models, but now there are more interesting questions in how we can adapt and control them for the specific workflows or problems for a given domain.

What are the best areas to scale automation in human-centered AI?

AI is already pretty prevalent in the workplace. As I write this spelling and grammar checkers as well as text autocomplete are helping me. I have spam filters and message classifiers on my email/messaging tools. I use AI-powered search to find the relevant information I need to execute. This will grow as well as my adoption as the number of AI-powered features and their quality increases.

However, I would call this kind of AI augmentation atomic. It is certainly assisting me, but in very precise moments that allow for high-precision predictions. I certainly cannot ask an AI to answer these questions for instance — yet!

More seriously, my vision is to see the adoption of end-to-end AI throughout the workspace. I don’t mean end-to-end in the machine learning modeling sense. What I mean is that the AI will power holistically large and complex tasks being optimized for the overall goal and not just atomic points during the process. ASAPP is already bringing this to bear in call centers. For instance, we optimize what the agent will say next based on a holistic set of factors about where the agent is in the conversation and what the ultimate goal is. But beyond that, imagine a scientist trying to write a systematic review of an important topic, a software engineer building a platform or integrating complex systems, a lawyer writing a legal brief, etc. In the future, each of these professionals will rely on AI to rapidly increase their effectiveness at these tasks and optimize desired outcomes, freeing them up for more critical challenges.

Do you have some final thoughts?

Our research team at ASAPP has a clear focus: we’re advancing AI to augment human activity to address real-world problems for enterprises. Researchers at ASAPP work to fundamentally advance the science of NLP and ML toward our goal of deploying domain-specific real-world AI solutions, and to apply those advances to our products. They leverage the massive amounts of data generated by our products, and our ability to deploy AI features into real-world use to ask and address fundamental research questions in novel ways.

Discover our recent papers at https://www.asapp.com/ai-research/

About Ryan McDonald

Ryan McDonald is the Chief Scientist at ASAPP. He is responsible for setting the direction of the research and data science groups in order to achieve ASAPP’s vision to augment human activity positively through the advancement of AI. The group is currently focused on advancing the field of task-oriented dialog in real world situations like customer care. Ryan has been working on language understanding and machine learning for over 20 years. He has published over 100 research papers in top tier journals and conferences which have been cited thousands of times. He has won best paper awards at premier international conferences (EMNLP, NAACL) for his work on multilingual syntactic analysis. His book ‘Dependency Parsing’ has served as one of the main pedagogical resources in syntactic parsing for over a decade.

About ASAPP

ASAPP is a research-based artificial intelligence software provider that solves large, complex, data-rich problems with AI Native® technology. Large enterprises use ASAPP to make customer experience teams highly productive and effective by augmenting human activity and automating the world’s workflows. The company has offices in New York, Silicon Valley, Buenos Aires, London, and Bozeman. Visit www.asapp.com for more information.

The post aiTech Trend Interview with Ryan McDonald, Chief Scientist at Asapp first appeared on AITechTrend.

The History of Artificial Intelligence

Although today, we have artificial intelligence, its early history is so far gone, it will be difficult to trace it. The author says that there is no record of the first successful AI research project. However, it is known that before Google, Apple and other major companies made significant advances in AI, they used “black box” AI programs such as the ones found in IBM’s Deep Blue and Deep Blue West, which relied on pre-programmed chess or checkers rules. These programs were ultimately defeated by human players. In the 1960s, AI pioneer Marvin Minsky set up the first computer programming language. The language was called LISP, or Generalized Liar-like Language, and it was designed to help programmers build autonomous systems.

AI and The Future

Universities and big companies have been focusing on the development of AI technology. Silicon Valley, in particular, is filled with tech companies all trying to outdo each other in the race to develop the best AI.

We often hear the term ‘new world order, and AI is poised to take over all the jobs in order to create a new economy. But with all the hype around AI, what is it exactly?

Is it just a buzzword for making more money? Can robots become our masters? Or are they the servants who help us with everyday tasks? I would like to think that AI will make life easier. I would like to think that it will help us create a better and more sustainable world.

What is Artificial Intelligence?

In the beginning, the search for artificial intelligence was very basic. You had algorithms and even vacuum tubes that did not respond to human interactions and made it impossible for humans to do many tasks. Today, artificial intelligence is now a more technical thing and the chances of robots replacing all human workers are very low.

What Can We Expect From the Future of AI?

As a result of the huge growth in Artificial Intelligence startups, it’s not only the students who have become enthusiastic about it. The big enterprises are also looking at AI, especially after this year’s announcement of Google Duplex, where a machine learns from real-time conversations to be a good salesperson. The advancement in the form of Artificial Intelligence has been increasing in 2021 and people are wondering how they will take it further in the near future. However, before the machine could start out on its journey to getting smarter, the questions remain about whether this will mean more jobs will be created or it will become a digital freeloader. As an alumnus of the University of Waterloo and a tech investor, I find that the latter is more likely.

Conclusion

The future of technology is scary and exciting. Technology is no more just a computer or a phone, but it has become an entire ecosystem that is contributing to our well-being, by aiding in creating new possibilities in diverse fields of the world.

The post The Future of Technology and Artificial Intelligence: What Will Happen? first appeared on AITechTrend.