The paper trail automation with the help of AI

Rather than conceding to the fine, Oleksandr saw an opportunity to put his technical skills to the test. He wrote a script that took advantage of Google’s image-to-text API, converting the image of his speeding ticket into a textual format. Subsequently, the script leveraged the ChatGPT API’s capabilities to understand the purpose behind each field of the TR-205 form, a California legal form for contesting citations, and to fill it in automatically with the relevant data.

Oleksandr elaborated on the process:

“I developed a script that used Google’s image-to-text API to transform the photo of my speeding ticket into editable text. Then, I employed the ChatGPT API to discern the intent of the TR-205 form fields and to autonomously populate the form with accurate information.”

This innovative application of AI tools significantly streamlined the process, allowing Oleksandr to not only submit the form accurately but also to craft a detailed and persuasive argument for reducing his fine.

Was it worth it?

Indeed, it was. The fine was ultimately reduced by 75%.

“Today brought the court’s response. Although I was still found guilty, the fine was reduced to $110,” said Oleksandr.

This reduction was the result of his clear and compelling written declaration and meticulous attention to detail in filling out the TR-205 form. In the past, such tasks would require the costly services of specialized lawyers. Now, AI can handle them in a fraction of the time.

A pioneering approach with AI

As the founder of BOTMAKERS, one of the pioneering companies to receive access to the ChatGPT API, Oleksandr had a strong foundation for this endeavor. His company has already created over 25 ChatGPT plugins, displaying Oleksandr’s extensive experience with the ChatGPT API and its integration.

Harnessing this expertise, Oleksandr developed a straightforward application named InstaFill. This new app, available for free, simplifies the process of filling PDF forms by utilizing uploaded information. Building on this innovation, Oleksandr aims to expand its capabilities by developing a ChatGPT plugin that offers the same functionality.

Why it matters?

Oleksandr’s story sheds light on the potential of AI as a tool for legal assistance. It demonstrates that one doesn’t need to be a legal expert or a proficient English speaker to manage the intricacies of the legal system. AI technology, particularly the ChatGPT API, is equipping individuals with the ability to tackle legal challenges head-on, filling out complex forms, and ensuring their arguments are heard.

In our fast-paced and ever-changing world, technology is the key to turning unfamiliarity into empowerment. Oleksandr’s success clearly indicates that with the aid of modern tools, anyone can navigate through legal obstacles and secure a fair outcome. His experience serves as an inspiration to embrace AI and its burgeoning role in everyday challenges. The future of AI is not just on the horizon—it is here, offering solutions and support for those ready to explore its vast capabilities.

In June 2023, AIN.Capital informed that two ChatGPT plugins, developed by a Ukrainian team led by Oleksandr Gamaniuk, were included in OpenAI‘s official plugin catalog.

For the project, SoftServe is looking for Data Science specialists in Ukraine, Romania, Poland, Colombia, Spain, and Chile.

• The aim of the lab is to accelerate the implementation of disruptive AI technology through the adoption of GenAI in various fields and industries, helping businesses solve specific problems and implement innovations in the real world. It also aims to expand the company’s expertise in machine learning and multimodal AI.

“Unlike other areas, GenAi stands out with its novelty and emerged properties unveiling not just technological innovations, but also groundbreaking use cases that may redefine business strategies. The pace of innovation here is staggering, ofter overwhelming business with new advancements released weekly. This is where the Gen Ai Lab shines. It’s not just a hub for rapid experimentation, value discovery, and solution engineering with the latest techniques and technologies. It’s a conduit for channeling this explosive innovation directly to our clients”.

Alex Chubay, CTO, SoftServe, says.

• Within the Gen AI Lab, SoftServe will work in the following key areas:
  • research and validation of generative AI applications (AI Discovery),
  • experiments and testing of AI on specific business cases (AI Launchpad),
  • rapid scaling and industrialization of AI (AI Adoption).
• As part of the work in the Gen AI Lab, SoftServe specialists will use tools and technology platforms in partnership with AWS, Google Cloud, Microsoft Azure, and NVIDIA.

“SoftServe is always at the forefront of tech progress. Today, GenAI is a truly breakthrough technology that has the potential to radically transform both the technology sector and open new horizons in many industries. In our lab, we analyze and adapt this technology in depth, exploring its capabilities in software and Big Data processing for specific business areas and industries. That’s why we invite specialists who want to work on the latest technologies from the idea to the actual implementation of innovations, opening up new opportunities for our clients and partners,”

Yurii Milovanov, AVP AI & Data Science, SoftServe, comments.

About SoftServe

SoftServe is the largest global IT company founded in 1993 in Lviv by Ukrainians Taras Kytsmey and Yaroslav Lubinets. The company provides software development and consulting services. More than 12,000 employees work in more than 50 offices located from San Francisco to Singapore. The company’s headquarters are located in Lviv and Austin (USA). SoftServe’s clients include IBM, Cisco, Panasonic, Cloudera, Henry Schein, Spillman Technologies, and others.

About the AISDR

AiSDR automates sales prospecting using generative artificial intelligence to write and send emails to prospects.

AiSDR will cost significantly less than a full-time sales development representative but will work with your leads 24/7. The system will:

• create personalized email campaigns for each prospect
• send personalized follow-ups to prospects
• keep corresponding and engaging prospects to make an appointment with you.

According to the founders, this will help companies expand their reach without increasing headcount while giving them the freedom to experiment with campaigns before hiring their first SDR.

After only two months, AiSDR has already got 24 clients. The response rate to automated emails is similar to the industry’s rates for emails written by humans.

About the founders

In 2017, brothers Yuriy and Oleg Zaremba founded Axdraft, a legal document automation startup. In April 2019, AxDraft closed a $1.1 million funding round. During the round, the project was valued at $10 million.

In December 2020, the international LegalTech company Onit acquired AXDRAFT. After the sale, Yuriy Zaremba remained at the startup as Managing Director. However, the brothers have now left the startup to start a new project.

How to use Uncreative agency

In order to generate ideas, you need to input some details about what you need. Precisely, the brand you are promoting, what type of campaign it should be, who are you targeting, and what results willing to achieve. We ordered a TV commercial for AIN.Capital, that will make entrepreneurs subscribe to our site.

After submitting, the magic begins. It takes several minutes for AI to generate ideas, meanwhile, the client is watching it “drinking coffee, reading Reddit, watching sand, calling freelancer” and other stuff real creative people usually do (probably) while working on the project.

But here we’re dealing with algorithms, remember? So the project is ready within minutes and emailed to you as a PDF. You can check out what we’ve got in a slideshow below:

As the Uncreative agency website claimed, all the creative work is done by machines. It is also yet unclear who made the tool itself. All that is known, so far, is that the company is headquartered in California.

While Uncreative is in beta, the service is completely free. After that, according to the site, users will be charged $5 dollars per idea.

Why is it interesting

This tool was launched as a fun experiment and obviously can’t replace people, at least yet. The ideas it generates are quite common and superficial but can be useful for stressed marketers who are struggling to produce many ideas in a short period of time. Sometimes, rethinking and developing the idea is easier than creating it from scratch.

About the service

Which Face Is Real is an online service offering users to guess which of the two proposed photos is real and which is generated by artificial intelligence.

As the developers note, the service is based on StyleGAN, an algorithm developed by the Nvidia team. It pits two neural networks — one creates deepfakes, and the other tries to determine the fake. In this way, they learn from each other.

Which Face Is Real also posts a couple of images: a photo from the FFHQ collection and a synthetic one created by StyleGAN. In this way, users can learn to distinguish authentic pics on their own.

How to spot a deepfake

The site’s authors determine some of the common problems in a special section of the service. For example, one of the distinguishing features of the StyleGAN algorithm is that it usually creates shiny blobs that look like water spots on old photos. They can appear anywhere in the image, but often they show up at the interface between the hair and the background.

Besides, it is extremely difficult for a neural network to render hair realistically. Sometimes strands can be disconnected on the face or elsewhere. Also, artificial Intelligence can depict some parts of the hair with a watercolor effect.

It is also very difficult for algorithms to generate realistic eyeglasses. A common problem is an asymmetry: often, the frame will have one style at the left and another at the right. In other cases, it will simply be jagged. In addition to the eyeglasses, you should pay attention to the asymmetries of facial hair, blurred accessories (such as earrings), and the different shapes of collar or fabric on the left and right sides.

People use AI-enabled technologies in everyday life — from booking flights and loan applications to driverless vehicles, medicine and diagnostics, and even decision-making by national governments. However, according to UNESCO, these technologies are also bringing unprecedented challenges.

“We see increased gender and ethnic bias, significant threats to privacy, dignity, and agency, dangers of mass surveillance, and increased use of unreliable AI technologies in law enforcement. Until now, there were no universal standards to provide an answer to these issues,” reads the UNESCO statement.

So, the document adopted by UNESCO member states is supposed to create legal infrastructure for the development of AI technologies within the ethical framework. It establishes a global normative framework for the industry and gives states the responsibility to apply it at their level. This is the first document of its kind. You can read the document here.

AIN.UA talked to Mariya Ortynska, patent attorney and CEO of IPStyle, about the legal implications of this document and how it will affect the work of developers who create AI solutions.

Experts have been discussing the ethical and legal regulation of AI for years. For example, can we qualify AI as an inventor on a par with humans or not? As in the case of DABUS, for which the first patent has already been granted.

Or as another example — we are now witnessing and even participating in the active development and introduction of driverless transport. But suppose we have a situation where autopilot malfunctions. Who is responsible for safety? Who is liable for the AI error and causing harm to human life and health?

As of today, there is no consensus, no single answer to these and similar questions. But there is no doubt that the time has come to discuss the future of artificial intelligence policy at the international level. In fact, the document adopted at the UN level gives rise to a normative legal framework regulating the development of AI and the ethics of its use. It also holds states responsible for its application at the national level. Moreover, UNESCO will ask states to report regularly on their progress and practices.

Transparency, clarity, and ethics will mean that soon AI will not be a “loot box” that can realize the scariest scenarios of sci-fi books or movies but will work as a quite understandable tool for the user to operate services and devices. For developers, these recommendations can be a guide when developing an AI product concept. Developers have global plans for AI.

Thus, the global market for explainable artificial intelligence (XAI) is estimated to be $3.50 billion by 2020 and is projected by Research And Markets to reach $21.03 billion by 2030. Explainable artificial intelligence is an emerging field of computer science, which forms AI models that can be better understood by humans — for example, how a program makes a decision, what the decision criteria are. Often not only the end-users but also the developers themselves cannot determine exactly how the model came to certain conclusions during data processing.

UNESCO itself relies on data showing that China and North America will have the greatest economic growth by 2030, making up 70% of the global economic impact of AI. The concentration of AI in the hands of a few high-income countries is likely to leave developing countries behind. The latter will have little or no benefit from AI and will not own such technologies.

According to the report, in Ukraine there are 57 companies that operate in the field of AI. Only Poland (110) and Russia (133) have more companies in the region. At the same time, there are only 11 industry investors in Ukraine – this is the fifth indicator.

The authors of the study, referring to the Clutch.co database, note that Ukraine leads in the number of outsourcing companies in the field of artificial intelligence not only in Eastern, but also in Western Europe. In total, there are 26 such institutions in the country, and 226 worldwide. The authors of the report emphasize that according to LinkedIn, there are more than 2,000 AI-development companies in Ukraine.

Most Ukrainian companies in the field of artificial intelligence are working on software (38%). Other popular areas include information technology, chatbots and AI assistants, fintech solutions and entertainment products.

Earlier AIN.UA published a guide to the IT sector of Ukraine from the UNIT.City, as well as a list of 40 more companies that could be there.

LUN began as an advertising search engine: it collected and processed information from other sites. “We use various kinds of information processing algorithms starting from the first day of the project’s existence, that is, for more than 10 years. Back in 2008, we implemented the first ad analysis system that used simple “if that” rules to find duplicate ads,” says Volodymyr Kubitskyi, the head of the team that works on AI-solutions. Fully neural networks were started to be introduced in LUN only in 2014, since then its number has exceeded two dozen. To support existing neural networks and work on new AI/ML-algorithms in the company, they created the Flatfy & Lun Artificial Intelligence Researchers team, FLAIR.

Outdoor advertising AI

They remembered about the developments of the team only this summer. In June, the company conducted an advertising campaign called “Do they build or only advertise?”. It promoted a new feature on the site – checking the reliability of the developer when buying an apartment in a new building. Within the framework of the advertising campaign, advertising boards were also planned, and it was necessary to find locations for them. According to Andriy Mima, co-founder of LUN, usually in such cases advertising agencies work according to one pattern: they send a list of several hundred selected advertising spots from which you need to choose, for example, 50. “I call it an “advantage”, like you can choose to your own taste.  And to make a choice, you only have the address of the board, its old low-quality photo and a strange OTS figure. And if you ask to make a choice for you, then they choose the most beautiful ones “by photo”— says Mima.

The company has decided to act differently and asked the agency to shoot a video with the boards, so that it was possible to determine whether the advertisement was clearly visible, whether tree branches or other boards were not obscuring. Then the company created a table with parameters for each board, which would make it easier to decide on each location. The main parameters for each board were identified as:

• board size;
• real picture (that shows whether the trees obscure the sign);
• time of non-visibility;
• competition with other boards in sight.

This label was created by viewing the video prepared by the agency, and manually writing out all the numbers. As a result, they managed to select 50 boards from the proposed 500 ones. However, the team had to spend several weeks on watching the video and filling in the table. Therefore, the next time the process was decided to automate:

• in order to analyze any number of ad slots,
• to increase the speed up such analysis.

At this stage, work with neural networks began. Just at the same time, the FLAIR team worked on a system for determining watermarks in photos of apartments, in order to select pictures without such signs for the title photo in search results. As a result, a system of localization of the object in the photo was created: neural networks showed several examples of what they would like to find in the photo, and then it learned to search basing on the new data itself.

“Just at that moment, the marketing guys were busy selecting sites for boards. We saw each other in the corridor, someone spoke a word about how they learned to detect watermarks and can generally find any object on any photos. So, the idea was to try to analyze the video from the registrar to find the boards,” says Volodymyr Kubitskyi.

According to him, the solution to search for advertising space was created from scratch. “For the implementation of the neural network, the tensorflow framework from Google was used to simplify the solution of the localization problem. In the transition from localization to tracking (localization is single-frame, and the board is contained in many video frames and should not be lost) they wrote their own post-processor,” he says.

How a neural network works

As a result, they managed to create a neural network that can accept video of almost any quality, in any light and at any shooting angle, and determine all ad slots on it, calculate the specified parameters for them: size, visibility time and competing boards that are located beside. This data is entered into a table, and then the marketing team can select any data evaluation formula and filter the results by it. For example, in the summer campaign, the formula looked like this:

Effect = Board area x Time in visibility zone / Number of other boards on the screen.

According to Kubitskyi, the neural network works this way:

• The input video is streamed into a set of frames (usually 24 frames per second) using OpenCV.
• Then the neural network begins to work with the Faster R-CNN architecture, trained to find the boards on the frames: the network gives the answer for each frame in the form of the coordinates of the found boards, if any.
• After this, the post-processing stage begins, where you need to reassemble the video from the frames and understand that the same board is tracked from frame to frame, and a new board is not tracked every time. Here, computer vision algorithms are connected to compare images and a Kalman filter for working with the found coordinates from frame to frame. Since the neural network has learned to work with multiple boards simultaneously, it calculates whether competing boards are in sight.
• All of this data is summarized in the report

According to Andriy Mima, the capabilities of such system can be broader than the current prototype. For example, you can show specific boards to the system and it will automatically find them on video, select them and make an assessment on the parameters. This can be useful for monitoring ongoing ad campaigns.

“I read somewhere that there are 1,600 km of roads in Kyiv. With an average speed of 16 km/h, you can drive around all of them with 3 cars in 4-5 working days, and if you remove the small roads, then, most likely, this task will still be a couple of times easier and cheaper. However, what to do with 100 hours of video then? It is 360,000 seconds, where you need to stop every 5-10 seconds, note the time, count the area and write it down in the table? It is necessary to measure about 50,000 structures, and it seems that this can take several months. It’s a huge amount of work that we have just automated,” says the co-founder of the company.

If it turns out that there is a demand for such technology among advertising agencies or sales houses, the company is ready to sell or license it.