Tag: chatbot

Snapchat brings its AI chatbot to all users, empowers more creators

[ad_1]
New Delhi: Snapchat has announced to roll out its AI chatbot called ‘My AI to all users globally with new features for a better way to chat with your friends.
Snapchat+ subscribers are currently sending nearly 2 million chat messages per day via ‘My AI’ to learn more about movies, sports, pets, and the world around them.
“Your AI comes with one of thousands of unique Bitmoji variations and can be easily customised to make it truly your own. Design a custom Bitmoji for your AI, give it a name, and start chatting,” said the company.

It’s now easy to bring My AI into any of your conversations with friends.
“My AI surfaces place recommendations from the Snap Map and
suggests relevant Lenses,” Snapchat, which has 750 million monthly active users, said during an event late on Wednesday.
Over 55,000 Snaps are now being created on Snapchat every second on average.
The company also announced other features on its main app.
Creators in an early pilot of the Stories revenue share programme are posting more often to Snapchat.

“Now, creators who have at least 50,000 followers, 25 million monthly Snap views, and post at least 10 Stories a month may be eligible to join our Stories revenue share programme,” said the company.
Every month, more than 100 million Snapchatters keep in touch through voice and video calling.
“New Calling Lenses let you break free from the grid and appear together, in one frame, and soon, even play games and solve puzzles while you’re virtually face-to-face,” the company added.
Snapchat now has a new type of Story called ‘After Dark.’
“Next time you’re up late studying or hanging out, add to the After Dark Story. Come morning, see the Story unfold to debrief on the night,” said Snapchat.
On Snap Map, the company is adding a new location sharing option.
“On Snap Map and beyond, 1.7 billion Snapchatters show up as their Bitmoji. This year, we added shoppable fashion so your Bitmoji could not only look, but also dress just like you,” according to the company.
[ad_2]
#Snapchat #brings #chatbot #users #empowers #creators
( With inputs from www.siasat.com )

April 20, 2023
Google denies it copied ChatGPT to train its own AI chatbot Bard

[ad_1]
New Delhi: Google has denied the reports that it is copying Microsoft-owned OpenAI’s ChatGPT to train its AI chatbot called Bard.
A report in The Information claimed that OpenAI’s success “has forced the two AI research teams within Google’s parent, Alphabet, to overcome years of intense rivalry to work together”.
According to the report, citing sources, software engineers at Google’s Brain AI group are working with employees at DeepMind, which is a sibling company within Alphabet to develop software to compete with OpenAI.
“Known internally as Gemini, the joint effort began in recent weeks, after Google stumbled with Bard, its first attempt to compete with OpenAI’s chatbot,” the report claimed.
However, a Google spokesperson told The Verge that “Bard is not trained on any data from ShareGPT or ChatGPT”.
Meanwhile, Google has announced it is opening up access to its ChatGPT competitor “Bard” as an early experiment for users to collaborate with generative AI.
Early access to Bard has rolled out in the US and the UK, and the company said it will expand the access over time to more countries and languages.
Bard, like OpenAI’s ChatGPT and Microsoft’s Bing chatbot, is based on a large language model (LLM), specifically a lightweight and optimised version of LaMDA, which the tech giant said will be updated with newer, more capable models in the future.

Users can interact with Bard by asking questions and refining their responses with follow-up questions.
[ad_2]
#Google #denies #copied #ChatGPT #train #chatbot #Bard
( With inputs from www.siasat.com )

March 30, 2023
GPT-4: OpenAI’s latest innovation in chatbot technology

[ad_1]
OpenAI, the leading artificial intelligence research institute, has announced the release of GPT-4, the latest version of its popular AI chatbot, ChatGPT. The new model boasts several enhanced capabilities that make it stand out from its predecessor, including the ability to process up to 25,000 words, eight times more than ChatGPT.
One of the most notable features of GPT-4 is its ability to respond to images. Users can upload photos of ingredients, and GPT-4 will provide recipe suggestions based on the image. Additionally, it can write captions and descriptions of images with impressive accuracy.
Since its launch in November 2022, millions of people have used ChatGPT for various purposes such as writing songs, poems, marketing copy, computer code, and homework assistance. However, there are concerns that GPT-4 could take over many jobs currently done by humans.
OpenAI has addressed these concerns and spent six months on safety features for GPT-4. The model was also trained on human feedback, but OpenAI warns that it may still be prone to sharing disinformation.
Initially, GPT-4 will only be accessible to users with premium access to ChatGPT, which requires a monthly subscription fee of $20. The model is already powering Microsoft’s Bing search engine platform, and the tech giant has invested $10 billion into OpenAI.
In a live demo, GPT-4 generated an answer to a complicated tax query. However, there was no way to verify the accuracy of its response. Despite its enhanced capabilities, OpenAI has warned that GPT-4 is still not fully reliable and may “hallucinate” or invent facts or make reasoning errors.
GPT-4 is a type of generative artificial intelligence, using algorithms and predictive text to create new content based on prompts. OpenAI claims that GPT-4 has more advanced reasoning skills than ChatGPT, and it can find available meeting times for three schedules, among other things.
OpenAI has also announced new partnerships with language learning app Duolingo and Be My Eyes, an application for the visually impaired, to create AI chatbots that can assist users using natural language.

[ad_2]
#GPT4 #OpenAIs #latest #innovation #chatbot #technology
( With inputs from www.siasat.com )

March 15, 2023
ChatGPT fails UPSC exam by 30! AI Chatbot could only answer 54 out of 100 questions – TheNewsCaravan Newspaper

[ad_1]
No doubt that UPSC is considered as the toughest exam in India! Proving the same, even ChatGPT has failed to clear the UPSC exam. ChatGPT (Chat Generative Pre-trained Transformer) is designed to generate human-like articles based on the given input Reports suggest that ChatGPT failed in UPSC prelims exam. The AI chatbot could only answer 54 out of 100 questions from Question Paper 1 (Set A) of UPSC Prelims 2022. Considering last year’s cut-off, UPSC failed the exam.
OpenAI claims that ChatGPT has limited knowledge of world and events after 2021. AIM claims that they went ahead and sent ChatGPT all 100 questions from Question Paper 1 (Set A) from the UPSC Prelims 2022. Surprisingly AI chatbot ChatGPT could only answer 54 out of 100 questions.
AI Chatbot ChatGPT was asked questions from several topics such as Geography, Economy, History, Ecology, General Science and current affairs. Reports suggest that since the current affairs questions were also based on 2022, ChatGPT failed to answer. However, report also claims that ChatGPT provided wrong answers to topics from sections like General Science, Geography and Economy.
In another update ChatGPT, the AI tool has been banned by many institutes. ChatGPT, the chat friendly AI bot has taken over headlines for its quirky remarks, quick tools and for helping out students with their homework assignments. Central Board of Secondary Education, CBSE Board Exams are underway and the Board has issued a set of board exam guidelines. Board has prohibited the use of Artificial Intelligence-based ChatGPT.
Source: IndiaJobs247
[ad_2]
#ChatGPT #fails #UPSC #exam #Chatbot #answer #questions #TheNewsCaravan #Newspaper
( With inputs from : www.TheNewsCaravan.com )

March 7, 2023
Users can now change behaviour of AI Bing chatbot

[ad_1]
San Francisco, March 3 (IANS) Tech giant Microsoft has added a new feature to its artificial intelligence (AI)-powered Bing chatbot that allows users to choose different tones for responses.
Users can choose between three options for the chatbot’s responses — precise, balanced and creative.
The creative mode provides responses that are “original and imaginative”, whereas the precise mode gives “accuracy and relevancy for more factual and concise answers”, reports The Verge.
The tech giant has set the Bing chatbot’s default mode to balanced, which it thinks will strike the right balance between accuracy and creativity.
These new chat modes are currently rolling out to all Bing AI users, and about 90 per cent of users should already be seeing them.
According to Mikhail Parakhin, the head of web services at Microsoft, the update includes a “significant reduction in cases where Bing refuses to reply for no apparent reason”.
Last month, the company had said that it was also going to test an additional option that would allow users to choose the tone of the chat from more precise, to balanced and more creative.
“The goal is to give you more control on the type of chat behaviour to best meet your needs,” it added.

Subscribe us on
[ad_2]
#Users #change #behaviour #Bing #chatbot
( With inputs from www.siasat.com )

March 3, 2023
Google testing ChatGPT-like products, including chatbot Apprentice Bard

[ad_1]
San Francisco: Google is testing ChatGPT-like products, which use its LaMDA (Language Model for Dialogue Applications) technology, media reports said.
Under new AI-powered chat products, the tech giant is testing a chatbot called “Apprentice Bard,” where employees can ask questions and receive detailed answers similar to ChatGPT, also a new search desktop design that could be used in a question-and-answer form, reports CNBC, citing sources.
The testing follows an all-hands meeting where employees expressed concerns about the company’s competitive edge in AI, given the sudden popularity of ChatGPT, created by OpenAI, a Microsoft-backed startup.
Moreover, Google’s parent company Alphabet is working on a project under its cloud unit called Atlas, which is a “code red” response to ChatGPT, the large-language chatbot that took the public by storm late last year, said the report.
Further, it mentioned that Apprentice Bard, which looks similar to ChatGPT, includes a feature in which answers can include recent events, which ChatGPT doesn’t have yet.
“As a result of ChatGPT, the LaMDA team has been asked to prioritise working on a response to ChatGPT,” read one internal memo viewed by CNBC.
“In the short term, it takes precedence over other projects,” the email said.
Apprentice Bard took the place of “Meena”, a previous version of a smart chatbot that had been launched internally but was later discontinued, said the report.

Meanwhile, Microsoft announced a multi-year, multi-billion dollar investment in OpenAI, the developer behind AI-driven ChatGPT which has become a rage.The tech giant, which infused $1 billion in OpenAI in 2019, did not divulge the investment amount in the “third phase of our long-term partnership with OpenAI”. Earlier reports had claimed Microsoft may infuse up to $10 billion in OpenAI.
[ad_2]
#Google #testing #ChatGPTlike #products #including #chatbot #Apprentice #Bard
( With inputs from www.siasat.com )

February 1, 2023