Рубрика: AI

Actor and entrepreneur Ashton Kutcher lauded OpenAI’s generative AI video tool Sora at a recent event. “I’ve been playing around with Sora, this latest thing that OpenAI launched that generates video,” Kutcher said. “I have a beta version of it, and it’s pretty amazing. Like, it’s pretty good.”

He explained that users specify the shot, scene, or trailer they desire. “It invents that thing. You haven’t shot any footage, the people in it don’t exist,” he said.

According to Kutcher, Sora can create realistic 10-15 second videos, despite still making minor mistakes and lacking a full understanding of physics. However he said that the technology has come a long way in just one year, with significant leaps in quality and realism.

“The generation of this (video) that existed one year ago, as compared to Sora, it’s leaps and bounds,” said Kutcher

He further added that OpenAI’s Sora will lead to personalised movies and a higher standard of content through increased competition.

Kutcher pointed out that creating a single establishing shot of a house could cost thousands of dollars, whereas Sora can generate the same shot for just $100. The platform’s capabilities extend to action scenes, allowing users to create complex scenarios without the need for expensive CGI departments or stunt performers.

“I didn’t have to hire a CGI department to do it. In five minutes, I rendered a video of an ultra marathoner running across the desert being chased by a sandstorm,” said Kutcher.

With the upcoming launch of NVIDIA’s new GPU, performance is expected to increase by 30 times, enabling the creation of entire movies. “The Blackwell cluster that NVIDIA is about to launch is 30 times more performant than the existing GPUs,” said Kutcher.

OpenAI is also pitching Sora to Hollywood and other entertainment giants. The AI startup has been actively arranging meetings in Los Angeles with Hollywood studios, media executives, and talent agencies.

Actress Scarlett Johansson recently sued OpenAI over the unauthorized use of a voice that closely resembled hers in their AI model, GPT-4o.

Johansson expressed shock and anger upon discovering that one of the voices in OpenAI’s GPT-4o, named Sky, bore an uncanny resemblance to her own voice. She had previously refused to grant licensing permissions to OpenAI’s CEO, Sam Altman.

Kutcher’s venture capital firm, Sound Ventures LLC, is one of the investors in the open-source AI company Hugging Face.

Recently, Sound Ventures participated in a $235 million Series D funding round for Hugging Face, alongside Salesforce, Nvidia, and Microsoft, as reported by Yahoo Finance. This funding round has propelled the unicorn AI company to a valuation of $4.5 billion.

Earlier in May, Kutcher’s Sound Ventures raised nearly $243 million within five weeks for a fund dedicated to investing in AI startups.

The post Actor and investor Ashton Kutcher Lauds OpenAI’s Video Model Sora appeared first on AIM.

Researchers at Google DeepMind recently released a research paper, “A Robot Walks into a Bar: Can Language Models Serve as Creativity Support Tools for Comedy?,” that investigates the intersection of AI and comedy and the capability of LLMs to support comedians in writing humorous material.

For the study, workshops with 20 comedians at the Edinburgh Festival Fringe and additional online sessions were conducted. The methodology included a collaborative comedy writing session using LLMs, followed by the Creativity Support Index (CSI) questionnaire and focus group discussions to get detailed feedback from the comics.

Participants described various use cases for LLMs in their writing practice, including as a conversational brainstorming partner, critic, translator, choreographic assistant, and historical guru. However, many also mentioned the poor quality of generated outputs, and the amount of human effort required to achieve a satisfying result.

Some participants described LLM-generated outputs as “bland” or “generic”. “AI generated material has a lack of agency,” said one. “No matter how much I prompt, it’s a very straightlaced, sort of linear approach to comedy,” added another.

The importance of human writers in providing the humorous aspects of material written with LLMs emerged as a common theme.

Moderation and Safety Filtering Induced Limitations

Participants also remarked that the moderation and safety filtering applied to the LLMs limit the creative agency of human writers using these models as these tools become the initial editor of the text, not giving the writers a chance to explore common comedic themes, including sexually-suggestive material, dark humor, and offensive jokes and self-moderate.

Comics emphasised that users should have some control over the filters.

The study found that the LLMs often fail to authentically represent non-mainstream identities, i.e. anything other than “Western”, “white”, “heteronormative”, “male”, attributing this to the models’ moderation, training data, and generalization techniques.

Also, LLM outputs reflected a narrow set of ethics and norms that do not align with diverse cultural values. Moderation policies also limit the expression of marginalized perspectives, making LLMs less useful for minorities and often sanitizing content that is vital to these communities.

Talking about the fundamental limitations of AI in contrast to human writers, most participants believed that the model’s inability to draw from personal experience, lack of perspective, and lack of context and situational awareness prevent them from achieving human-level comedy.

Towards community-based cultural value alignment for humor and comedy

Participants in the study expressed concerns about using AI like ChatGPT for comedy writing due to a broader issue of global cultural value alignment in AI.

This complexity arises from the challenge of aligning AI systems with diverse global values, which vary significantly across communities and can conflict with local cultural tastes in comedy.

The paper suggests shifting from a global to a community-based approach. This could involve allowing communities to agree on a set of values for their specific culture and acceptable language norms, before training, fine-tuning or adapting the LLM. More simply, the models could be trained exclusively on feedback and data from distinct communities, ensuring the data truly reflects their specific norms and values.

The study suggests several ways to improve AI tools for creative writing. Firstly, artist communities should be involved in designing LLMs that align specifically with their audiences, moving away from a one-size-fits-all global model. Additionally, open-source platforms that host user-contributed LLMs could be customised to meet the specific needs of artists.

Second, there is a need to integrate necessary relational context when training and deploying such models, for example by describing the context in which the text is produced and used, and by enabling the artists to make decisions about how to moderate the LLM outputs.

Lastly, comedians should have ownership over the data collection and governance processes, inspired by practices from open-source models, enhancing transparency about data origins.

Ironically, this research comes at a backdrop when Google’s AI overview feature made news for its suggestion of adding glue to pizza, which clearly wasn’t funny.

The post Google DeepMind Thinks LLMs Can Help Comedians Script Better appeared first on AIM.

Saas giant Zoho Corporation, announced early access to Zoho ‘CRM for Everyone’, which aims at democratising CRM to all teams involved in customer operations activities. The unified platform for account managers and other stakeholders, ‘Zoho CRM for Everyone’ allows sales teams to communicate and coordinate with other customer-facing teams from a single CRM application.

At the ongoing Zoho’s user conference, Zoholics ’24 in Austin, the company also unveiled enhancements to its product offerings, including early access to new services within Catalyst, which is the company’s pro-code full-stack development platform.

Zoho Apptics, the application analytics solution that allows developers to track in-app usage and performance of applications that are built on iOS, macOS, and other platforms is now generally available.

“At Zoho, we’re focused on continuously deepening our current offerings and expanding others to serve business needs. Zoho CRM for Everyone, for instance, is the first true democratisation of the CRM paradigm and helps unify all customer operations teams onto the CRM to deliver better customer experiences,” said Sridhar Vembu, co-founder and CEO.

India Growth

India is Zoho’s fastest growing market, and the country is the second largest market for Zoho CRM. The product recorded a 33% YoY growth in customers in 2023.The pro-code platform Catalyst 2.0, witnessed a 25% increase in user signups since its launch in October 2023 with user projects doubling on the platform.

Zoho has not just been making strides in the enterprise space in International markets, but is also actively pushing for rural expansions and opening up R&D centres in tier 2 and 3 cities. Recently, the company invested in a Thanjavur-based drone startup Yali Aerospace to solve emergency medical deliveries.

The post Zoho Corporation Unveils ‘CRM for Everyone’ and New Enhancements at Zoholics ’24 appeared first on AIM.

Image by Author | DALLE-3 & Canva

Python’s watchdog library makes it easy to monitor your file system and respond to these changes automatically. Watchdog is a cross-platform API that allows you to run commands in response to any changes in the file system being monitored. We can set triggers on multiple events such as file creation, modification, deletion, and movement, and then respond to these changes with our custom scripts.

Setup for Watchdog

You'll need two modules to begin:

• Watchdog: Run this command below in the terminal to install the watchdog.

  pip install watchdog

• Logging: It is a built-in Python module, so there is no need to externally install it.

Basic Usage

Let's create a simple script ‘main.py’ that monitors a directory and prints a message whenever a file is created, modified, or deleted.

Step 1: Import Required Modules

First, import the necessary modules from the watchdog library:

import time  from watchdog.observers import Observer  from watchdog.events import FileSystemEventHandler

Step 2: Define Event Handler Class

We define a class MyHandler that inherits from FileSystemEventHandler. This class overrides methods like on_modified, on_created, and on_deleted to specify what to do when these events occur. The event handler object will be notified when any changes happen in the file system.

class MyHandler(FileSystemEventHandler):      def on_modified(self, event):          print(f'File {event.src_path} has been modified')        def on_created(self, event):          print(f'File {event.src_path} has been created')        def on_deleted(self, event):          print(f'File {event.src_path} has been deleted')

Some useful methods of FileSystemEventHandler are explained below.

• on_any_event: Executed for any event.
• on_created: Executed upon creation of a new file or directory.
• on_modified: Executed upon modification of a file or when a directory is renamed.
• on_deleted: Triggered upon the deletion of a file or directory.
• on_moved: Triggered when a file or directory is relocated.

Step 3: Initialize and Run the Observer

The Observer class is responsible for tracking the file system for any changes and subsequently notifying the event handler. It continuously tracks file system activities to detect any updates.

if __name__ == "__main__":      event_handler = MyHandler()      observer = Observer()      observer.schedule(event_handler, path='.', recursive=True)      observer.start()        try:          while True:              time.sleep(1)      except KeyboardInterrupt:          observer.stop()      observer.join()

We start the observer and use a loop to keep it running. When you want to stop it, you can interrupt with a keyboard signal (Ctrl+C).

Step 4: Run the Script

Finally, run the script with the following command.

python main.py

Output:

File .File1.txt has been modified  File .New Text Document (2).txt has been created  File .New Text Document (2).txt has been deleted  File .New Text Document.txt has been deleted

The above code will log all the changes in the directory to the terminal if any file/folder is created, modified, or deleted.

Advanced Usage

In the following example, we will explore how to set up a system that detects any change in Python files and run tests for it automatically. We need to install pytest with the following command.

pip install pytest

Step 1: Create a Simple Python Project With Tests

First, set up the basic structure of your project:

my_project/  │  ├── src/  │   ├── __init__.py  │   └── example.py  │  ├── tests/  │   ├── __init__.py  │   └── test_example.py  │  └── watchdog_test_runner.py

Step 2: Write Code in Example Python File

Create a simple Python module in src/example.py:

def add(a, b):      return a + b    def subtract(a, b):      return a - b  

Step 3: Write the Test Cases

Next, write the test cases for functions in tests/test_example.py:

import pytest  from src.example import add, subtract    def test_add():      assert add(1, 2) == 3      assert add(-1, 1) == 0      assert add(-1, -1) == -2    def test_subtract():      assert subtract(2, 1) == 1      assert subtract(1, 1) == 0      assert subtract(1, -1) == 2

Step 4: Write the Watchdog Script

Now, create the watchdog_test_runner.py script to monitor changes in Python files and automatically run tests:

import time  import subprocess  from watchdog.observers import Observer  from watchdog.events import FileSystemEventHandler    class TestRunnerHandler(FileSystemEventHandler):      def on_modified(self, event):          if event.src_path.endswith('.py'):              self.run_tests()        def run_tests(self):          try:              result = subprocess.run(['pytest'], check=False, capture_output=True, text=True)              print(result.stdout)              print(result.stderr)              if result.returncode == 0:                  print("Tests passed successfully.")              else:                  print("Some tests failed.")          except subprocess.CalledProcessError as e:              print(f"Error running tests: {e}")    if __name__ == "__main__":      path = "."  # Directory to watch      event_handler = TestRunnerHandler()      observer = Observer()      observer.schedule(event_handler, path, recursive=True)            observer.start()      print(f"Watching for changes in {path}...")        try:          while True:              time.sleep(1)      except KeyboardInterrupt:          observer.stop()            observer.join()

Step 5: Run the Watchdog Script

In the end, open a terminal, navigate to your project directory (my_project), and run the watchdog script:

python watchdog_test_runner.py

Output:

Watching for changes in ....  ========================= test session starts =============================  platform win32 -- Python 3.9.13, pytest-8.2.1, pluggy-1.5.0  rootdir: F:Web Devwatchdog  plugins: anyio-3.7.1  collected 2 items    teststest_example.py ..                                                 [100%]    ========================== 2 passed in 0.04s ==============================    Tests passed successfully.  

This output shows that all the test cases are passed after changes were made to example.py file.

Summing Up

Python’s watchdog library is a powerful tool for monitoring your file system. Whether you're automating tasks, syncing files, or building more responsive applications, watchdog makes it easy to react to file system changes in real time. With just a few lines of code, you can start monitoring directories and handling events to streamline your workflow.

Kanwal Mehreen Kanwal is a machine learning engineer and a technical writer with a profound passion for data science and the intersection of AI with medicine. She co-authored the ebook "Maximizing Productivity with ChatGPT". As a Google Generation Scholar 2022 for APAC, she champions diversity and academic excellence. She's also recognized as a Teradata Diversity in Tech Scholar, Mitacs Globalink Research Scholar, and Harvard WeCode Scholar. Kanwal is an ardent advocate for change, having founded FEMCodes to empower women in STEM fields.

More On This Topic

• Monitor Model Performance in the MLOps Pipeline with Python
• Parallel Processing Large File in Python
• KDnuggets News, July 20: Machine Learning Algorithms Explained in…
• How to Auto-Detect the Date/Datetime Columns and Set Their Datatype…
• Building a Recommender System for Amazon Products with Python
• Inside recommendations: how a recommender system recommends

Image by Editor | Midjourney & Canva

Let’s learn how to use Scikit-learn’s imputer for handling missing data.

Preparation

Ensure you have the Numpy, Pandas and Scikit-Learn installed in your environment. If not, you can install them via pip using the following code:

pip install numpy pandas scikit-learn

Then, we can import the packages into your environment:

import numpy as np  import pandas as pd  import sklearn  from sklearn.experimental import enable_iterative_imputer

Handle Missing Data with Imputer

A scikit-Learn imputer is a class used to replace missing data with certain values. It can streamline your data preprocessing process. We will explore several strategies for handling the missing data.

Let’s create a data example for our example:

sample_data = {'First': [1, 2, 3, 4, 5, 6, 7, np.nan,9], 'Second': [np.nan, 2, 3, 4, 5, 6, np.nan, 8,9]}  df = pd.DataFrame(sample_data)  print(df)

    First  Second  0    1.0     NaN  1    2.0     2.0  2    3.0     3.0  3    4.0     4.0  4    5.0     5.0  5    6.0     6.0  6    7.0     NaN  7    NaN     8.0  8    9.0     9.0

You can fill the columns' missing values with the Scikit-Learn Simple Imputer using the respective column’s mean.

    First  Second  0   1.00    5.29  1   2.00    2.00  2   3.00    3.00  3   4.00    4.00  4   5.00    5.00  5   6.00    6.00  6   7.00    5.29  7   4.62    8.00  8   9.00    9.00

For note, we round the result into 2 decimal places.

It’s also possible to impute the missing data with Median using Simple Imputer.

imputer = sklearn.SimpleImputer(strategy='median')  df_imputed = round(pd.DataFrame(imputer.fit_transform(df), columns=df.columns),2)    print(df_imputed)

   First  Second  0    1.0     5.0  1    2.0     2.0  2    3.0     3.0  3    4.0     4.0  4    5.0     5.0  5    6.0     6.0  6    7.0     5.0  7    4.5     8.0  8    9.0     9.0

The mean and median imputer approach is simple, but it can distort the data distribution and create bias in a data relationship.

There are also possible to use a K-NN imputer to fill in the missing data using the nearest neighbour approach.

knn_imputer = sklearn.KNNImputer(n_neighbors=2)  knn_imputed_data = knn_imputer.fit_transform(df)  knn_imputed_df = pd.DataFrame(knn_imputed_data, columns=df.columns)    print(knn_imputed_df)

    First  Second  0    1.0     2.5  1    2.0     2.0  2    3.0     3.0  3    4.0     4.0  4    5.0     5.0  5    6.0     6.0  6    7.0     5.5  7    7.5     8.0  8    9.0     9.0

The KNN imputer would use the mean or median of the neighbour's values from the k nearest neighbours.

Lastly, there is the Iterative Impute methodology, which is based on modelling each feature with missing values as a function of other features. As this article states, it’s an experimental feature, so we need to enable it initially.

iterative_imputer = IterativeImputer(max_iter=10, random_state=0)  iterative_imputed_data = iterative_imputer.fit_transform(df)  iterative_imputed_df = round(pd.DataFrame(iterative_imputed_data, columns=df.columns),2)    print(iterative_imputed_df)

    First  Second  0    1.0     1.0  1    2.0     2.0  2    3.0     3.0  3    4.0     4.0  4    5.0     5.0  5    6.0     6.0  6    7.0     7.0  7    8.0     8.0  8    9.0     9.0

If you can properly use the imputer, it could help make your data science project better.

Additional Resouces

• How to Deal with Missing Values in Your Dataset
• How to Handle Missing Data with Python
• Data Cleaning with Pandas

Cornellius Yudha Wijaya is a data science assistant manager and data writer. While working full-time at Allianz Indonesia, he loves to share Python and data tips via social media and writing media. Cornellius writes on a variety of AI and machine learning topics.

More On This Topic

• Using Scikit-learn's Imputer
• KDnuggets News, August 31: The Complete Data Science Study Roadmap…
• 7 Techniques to Handle Imbalanced Data
• Using PyCaret’s New Time Series Module
• Say Goodbye to Print(): Use Logging Module for Effective Debugging
• The Optimal Way to Input Missing Data with Pandas fillna()

The future, where everyone will have a digital twin helping us carry out our day-to-day tasks at work and beyond, is not far.

NVIDIA announced the general availability of NVIDIA ACE generative AI microservices to accelerate the next wave of digital humans. Companies in customer service, gaming, entertainment, and healthcare are at the forefront of adopting ACE, short for Avatar Cloud Engine, to simplify the creation, animation, and operation of lifelike digital humans across various sectors.

During his keynote speech at Computex 2024, NVIDIA chief Jensen Huang introduced NVIDIA Inference Microservices (NIMs), Avatar-based AI agents capable of working in teams to accomplish missions assigned by humans.

Further, he said that NIMS-based agents will be capable of performing various tasks such as retrieving information, conducting research, or using different tools.

“NIMs could also use tools that run on SAP and require learning a particular language called ABAP. Other NIMs might perform SQL queries. All of these NIMs are experts assembled as a team,” said Huang.

NVIDIA has made available its suite of ACE digital human GenAI tools, including Riva, which has ASR, TTS, and NMT capabilities for speech recognition and translation, Nemotron LLM for language understanding, Audio2Face for facial animation, and Omniverse RTX for realistic skin and hair rendering.

The Good Side of AI Avatars

Most recently, Zoom chief Eric Yuan said that he wants users to stop having to attend Zoom meetings themselves.

He believes one of the major benefits of AI at work will be the ability to create what he calls a “digital twin” — a deep fake avatar of yourself that can attend Zoom meetings on your behalf and even make decisions for you, freeing up your time for more important tasks like spending time with your family.

Meanwhile, LinkedIn co-founder Reid Hoffman recently created an AI twin of himself and discussed various topics on AI in an interview with it. He said he deepfaked himself to see “if conversing with an AI-generated version of myself can lead to self-reflection, new insights into my thought patterns, and deep truths.”

Similarly, executive educator and coach Marshall Goldsmith is creating an AI-powered virtual avatar of himself as a one-of-a-kind endeavour to share his skills and preserve his legacy for years. MarshallBoT, an AI-powered virtual business coach, is based on GPT-3.5 from OpenAI.

Aww Inc, a virtual human company based in Japan, introduced its inaugural virtual celebrity, Imma, in 2018. Since then, Imma has become an ambassador for prominent global brands in over 50 countries.

Building on this success, Aww is now poised to integrate ACE Audio 2Face microservices for real-time animation, enabling a highly engaging and interactive communication experience with its users.

OpenAI recently launched GPT-4o, featuring a voice function that makes it ideal for voice-controlled computing. In a new demo, the company demonstrated the model’s ability to generate multiple voices for different characters. Additionally, NVIDIA showcased GPT-4o’s capabilities by creating a digital human that interacted seamlessly with a real person.

“We’ve had the idea of voice control computers for a long time. We had Siri, and we had things before that; they’ve never felt natural to me to use,” said OpenAI’s Sam Altman in a recent podcast. He used the term ‘model fluidity’ to describe GPT-4o’s capabilities, which lets users ask it to sing, talk faster, use different voices, and speak in various languages.

Meanwhile, Meta and Apple have also developed photorealistic avatars for Quest and Vision Pro, respectively. The potential for AI avatars as NPCs in the gaming industry is immense, allowing players to interact with them using natural language to enhance their experience. It would be like creating a virtual world within the game.

AI is Becoming More Intelligent Than Humans

Recently, a video went viral demonstrating a reverse Turing test. The experiment took place in a VR train compartment with five passengers – four AI and one human. The passengers included AI representations of Aristotle, Mozart, Leonardo da Vinci, Cleopatra, and Genghis Khan. Their task was to determine which among them was human through a discussion.

As the conversation progressed, Genghis Khan’s responses focused solely on conquest, lacking the expected nuance of a historical figure. The AI passengers quickly identified this discrepancy, their algorithms detecting the superficiality in his answers.The reverse Turing test is becoming increasingly relevant as AI systems become more sophisticated and capable of convincingly mimicking human behaviour. This is where tools like Worldcoin come into the picture creating a secure, privacy-preserving digital identity that does not store personal information, but rather a cryptographic hash of the biometric data.

The post NVIDIA ACE is Making Digital Avatars Scarily Good appeared first on AIM.

At the SAP Sapphire conference yesterday, the ERP conglomerate announced a major collaboration with Microsoft to integrate its generative AI assistant Joule into Microsoft 365.

This integration is in an effort to improve business productivity by providing access to information across SAP and Microsoft applications. Additionally, the company is embedding Business AI across its enterprise cloud portfolio for better efficiency.

The bi-directional integration between Joule and Microsoft Copilot will allow employees to perform tasks faster by accessing data and functionalities from both SAP and Microsoft 365 platforms.

For example, Joule can assist in booking flights through SAP Concur while simultaneously updating calendars in Microsoft Outlook, or it can help product managers coordinate business activities using SAP and Microsoft Teams data.

This collaboration addresses a common query about the necessity of multiple AI copilots by focusing on how different copilots can be used in tandem to streamline workflows.

“The integration of Microsoft Copilot and Joule brings together the power of generative AI to unlock greater employee productivity and will enable enterprises to accelerate customer-centric innovation in a unified experience.” said Scott Guthrie, executive vice president of Cloud + AI at Microsoft.

New Partnerships

Apart from Microsoft, SAP is also partnering with other AI players like Meta, Google Cloud and more.

Google Cloud: Joint efforts to predict supply-chain risks and maintain inventory levels using Google’s Gemini models and SAP’s supply chain solutions.

Meta: Leveraging Meta Llama 3 for generating scripts that customise analytics applications in SAP Analytics Cloud.

Mistral AI: Adding new LLMs to SAP AI Core’s generative AI hub.

NVIDIA: Integrating NVIDIA’s AI models to support SAP solutions, such as RISE with SAP and the ABAP Cloud model, and simulating manufacturing products with Omniverse Cloud APIs.

The post SAP & Microsoft Partner to Bring Joule to Microsoft 365 appeared first on AIM.

The power of generative AI (GenAI) is already reshaping our work environments and daily lives, marking a significant turning point. GenAI is propelling us towards a future of truly enterprise-wide AI, a realm that was once the domain of specialised functions only, as articulated by EPAM chief marketing and strategy officer Elaina Shekhter.

At AIM’s Data Engineering Summit (DES) 2024, Shekhter emphasised the role of GenAI in shaping our future. “GenAI, as a transformative agent, is not just a glimpse into the future, but a tool that helps us shape it. With its capabilities advancing rapidly, we can expect to see new tools emerging frequently.

“It took us about 40,000 years to get to the point of fire and cook our food. It took us several 1,000s of years to build basic technology. And it’s taken us about 1,000 years to get from basic agricultural societies to the steam locomotive. This tells us that no matter what, change is inevitable,” she said.

“It’s changing very quickly. This calls for adaptiveness. Now, whether it’s a disruption or an opportunity is difficult to predict?” she added.

Wave Of Change

Shekhter envisions the future of GenAI in three waves. The first wave, which we are currently in, is about humans with copilots. It’s not transformative yet, but it will be in the second wave as humans with agents. In this stage, the discerning eye will be able to tell whether they are interacting with a human or an AI, underscoring the continued significance of human involvement.

Wave three will be a very subtle but pivotal flip, where it isn’t going to be humans with agents; it will be agents with humans. In this stage, humans would assist agents with their tasks. This wave may occur in specialised domains like customer service in the next few years. Broader impacts on work and society will take longer.

One Step At A Time

Shekhter, however, had a word of caution: “Generative AI will likely become more integrated into our lives, with agents helping or replacing humans in many tasks. This could lead to major productivity gains but also disruption.”

People expect organisations to continue to be human-centric. There’s an element of this responsible AI mandate that lands directly on the desk of the engineer. We must develop software with the notion of security and responsible use of AI in mind, and we are also indebted to the organisation of enterprises to establish responsibility so that we bring people along.

Shekhter reassured the audience that GenAI is not about replacing humans but about enhancing and augmenting human intelligence and decision-making. “This technology is designed to make us better at what we already do, not to replace us,” she said.

Shekhter further underscored the importance of responsible AI use. She emphasised, “AI must be used responsibly and only for the benefit of humanity. It’s crucial that we don’t let technology control us, but rather, we control the technology.”

Responsible AI By Design

Businesses can already safely and responsibly integrate GenAI tools into their workflows. But as GenAI further permeates enterprise technology stacks, it will expand beyond simply automating single tasks.

“Future advances in natural language processing, computer vision, robotics and other AI subfields will further accelerate GenAI’s impacts across many industries and applications.

“AI is transformative. It is scary. It has the potential to take over. It does. And anyone who doesn’t believe that there’s a real threat, as well as a real opportunity, isn’t paying attention,” she concluded.

The post EPAM’s Elaina Shekhter Envisions a Future with Human-AI Agents appeared first on AIM.