Connect with us

Tech

What is Gemini? Everything you need to know about Google’s latest AI model

Published

on

What is Gemini? Everything you need to know about Google's latest AI model

Google’s most recent generative AI model, Gemini, was just released. The well-known tech corporation regards Gemini as the most sophisticated and adaptable AI model. Google has said that it will expand and improve its large language model (LLM) in the next year. Because the LLM is multimodal, it can understand several data types, including text, audio, pictures, and video.

Google Gemini: What is it?

What is Gemini? Everything you need to know about Google's latest AI model

The most recent large language model (LLM), Google Gemini, created by Google AI’s DeepMind, has sophisticated multimodal processing skills. It can comprehend, work with, and combine many kinds of data, such as text, code, audio, pictures, and videos.

This potent AI has three variants: Ultra, Pro, and Nano. Each version is intended to tackle a certain task’s level of complexity. Gemini has outperformed benchmarks and current industry norms. Its versatility across platforms and adherence to moral AI standards—which have been put through rigorous safety and bias testing—set it apart. Google intends to include Gemini in its offerings, available via Google Cloud Vertex AI and Google AI Studio. Notably, the Gemini Pro edition is perfect for various jobs and is offered for free.

How is Google Gemini used?

What is Gemini? Everything you need to know about Google's latest AI model

How you use Google Gemini depends on the integrated product and version. With Google Bard, for instance, users can submit questions and get answers on everything from weather predictions to poetry writing to coding help—all while being protected from offensive material.

Gemini Nano connects with Gboard for Pixel 8 Pro users, offering recommended responses in messaging applications like WhatsApp. Moreover, Nano may summarize offline recorded talks in the Recorder app.

Although Gemini Ultra’s functionality is unknown, it seems designed for complex activities and may be aimed at academics and business users. Its projected inclusion with Google’s chatbot, Bard Advanced, promises further exploration of opportunities when it is released.

GPT 4 vs Google Gemini: What’s the difference?

What is Gemini? Everything you need to know about Google's latest AI model

With its remarkable multimodal processing powers, Google Gemini excels at managing text, code, audio, photos, and movies. Its ability to integrate and manipulate many kinds of data allows for flexible interactions across several modalities, which is its main strength. Conversely, GPT models like as GPT-3 and maybe GPT-4 do very well in natural language tasks, but they may not have Gemini’s broad capabilities to handle a variety of data formats outside of text. Gemini is unique in that it can handle jobs with a variety of data formats because to its skill with multimodal inputs, while GPT models are mostly used for text-based tasks.

Google displayed the results of eight benchmarks that included text, with Gemini coming out on top in seven. Google claims that Gemini won every single one of the ten multimodal benchmarks.

Though it’s not quite that simple, it would appear to suggest that Gemini is the better method. Gemini is effectively catching up to a nine-month-old #AI tool, since GPT-4 was released in March 2023. It’s difficult to determine which tool is now superior since we don’t know how powerful OpenAI’s next GPT version will be.

Furthermore, Google limited the competition for Gemini Ultra to GPT-4. This indicates that although the current state of Gemini Pro and Nano’s ability to compete with GPT-4 is unknown, OpenAI’s model most likely outperforms Gemini Pro and Nano given the often narrow margins separating GPT-4 and Gemini Ultra.

In summary

With its multimodal proficiency, Google Gemini proves to be a capable multitasker in a future where artificial intelligence fuels human interactions. Although it leads in most benchmarks, the competition with GPT-4 is still close, making the victor of the race still to be announced. The competition for dominance between the two systems is still ongoing as they develop, indicating a bright future for AI research.

Related VOR News:

Apple Announces Its Annual Developers Conference Is Set For June 10

Canuckle – Play Canadian Wordle Game Online

Unblocked Games 66 EZ: Ultimate Fun at your Fingertips

Continue Reading

Tech

Google Search Is Set To Undergo Significant Evolution By 2025. Sundar Pichai

Published

on

Google

(VOR News) – Sundar Pichai, the chief executive officer of Google, has stated that the search engine that the company employs would undergo a significant transformation by the year 2025.

This change will be made possible with the assistance of cutting-edge artificial intelligence. At the DealBook Summit, which was hosted by the New York Times, Sundar Pichai highlighted his faith in artificial intelligence as the primary driving force behind the shift. He said this during his presentation at the event.

According to the findings of the researcher, “Search itself will continue to undergo profound changes in the year 2025.”

Even if you wait until the beginning of the year 2025, I am confident that you will be astounded by the more advanced capabilities that search has in contrast to where it is at the moment. It is my firm belief that you will be astonished.

Google shared this information as part of its continuous efforts to incorporate artificial intelligence into its products. These efforts are ongoing.

Recent changes include search summaries that are generated by artificial intelligence as well as enhancements to its Lens feature, which now supports web searches that are based on videos. Additionally, the Lens feature has been included in the list of recent updates.

Another new development from Google is search summaries.

Additionally, in order to compete with competitors such as Microsoft and OpenAI, Google is presently in the process of creating a significant modification to its Gemini model.

This adjustment is being taken into consideration. The purpose of this adjustment is to enhance the company’s ability to compete with other businesses.

The audience was informed that artificial intelligence technologies such as BERT and MUM had already helped Google improve the accuracy and functionality of search results. Pichai made the audience aware of this information. He also said that more than one billion users use Gemini-powered features. This is an addition to the prior statement.

A few words were made by Satya Nadella, the Chief Executive Officer of Microsoft, in the beginning of this year. Sundar Pichai, who was there at the summit, responded to those remarks while he was there. According to Nadella, Google ought to have been the “default winner” in the fight for artificial intelligence.

He offered this suggestion. A number of statements were made by Pichai in response to the fact that Microsoft relies on OpenAI models. One of these statements was as follows: “I would love to do a side-by-side comparison of Microsoft’s own models and our models at some point in the future.”

Furthermore, Google uses models created by different individuals.

Additionally, the topic of conversation shifted to the compromise that needs to be made between fair usage and persons who create content getting compensated for their work.

This was the centre of the discussion. The ongoing discussion that is taking place regarding the value of intellectual property in the era of artificial intelligence was brought up by Pichai, despite the fact that he brought attention to the licensing arrangements that Google has with a variety of platforms, such as Reddit, the Associated Press, and The New York Times.

Pichai made it a point to mention the ongoing discussion. Sundar Pichai stated that a future marketplace would make it possible for content artists to create content that is specifically tailored for AI models and receive direct remuneration for their efforts.

He also stated that this opportunity would be available to them. Artificial intelligence would be included into the market, which would make this a possibility.

In his concluding remarks, Sundar Pichai emphasised the necessity of search in this day and age, when there is a lot of information available. He said that search is especially important in this day and age. He made the observation that “search becomes even more valuable in a world that is saturated with content.”

It is consistent with what he described that this is the case. Given that significant shifts are on the horizon, Google is placing itself in a position to continue to be a pioneer in the creation of technology that is driven by artificial intelligence. This is in light of the fact that substantial transformations are on the horizon.

SOURCE: TN

SEE ALSO:

FBI Warns iPhone and Android Users to Stop Texting

The US FDA Cited Musk’s Neuralink Animal Laboratory For ‘Objectionable Conditions.’

Continue Reading

Tech

FBI Warns iPhone and Android Users to Stop Texting

Published

on

FBI Warns iPhone and Android Users to Stop Texting

The FBI has warned Android users to refrain from texting messaging iPhone users and vice versa in a statement following a report that revealed text data was easily hackable due to a lack of encryption.

To keep others from listening in on their discussions, the FBI is advising users to use safe, encrypted messaging and phone calls using apps like WhatsApp. The warning follows news that Chinese hackers named “Salt Typhoon” had gained access to many US networks, with the extent of the breach perhaps being more than first believed.

‘Salt Typhoon’ is the gang that has already targeted the metadata of many clients, including information like call and text recipients, dates, and hours. A senior FBI official stated that “the facts will evolve over time within the investigative activity, especially one this significant and this large.”

The ongoing probe into the PRC’s targeting of commercial telecom infrastructure has uncovered a large-scale and noteworthy cyber espionage operation.

FBI Warns iPhone and Android Users

In addition to stating that “the FBI began investigating this activity in late spring and early summer of this year,” he cautioned that this effort “identified that PRC affiliated cyber actors have compromised networks of multiple telecom companies to enable multiple activities.”

“Using a cell phone that automatically receives timely operating system updates, responsibly managed encryption, and phishing resistant MFA for email, social media, and collaboration tool accounts” is what the FBI official advised individuals to do.

Adding to this, Jeff Greene of CISA told Forbes, “We definitely need to do that, kind of look at what it means long-term, how we secure our networks.” He strongly encouraged Americans to “use your encrypted communications where you have it.”

According to the information available, the Salt Typhoon attacks did not steal extensive call and text content, but the FBI official cautioned that extensive call and text metadata was. However, “a small number of people who are primarily involved in government or political activities had their private communications compromised by the actors.” This would have included text and phone contents.

Political upheaval is hardly surprising given the scope of the hacking campaign and the consequences for US critical infrastructure and network security.

Chinese Hacking Telecommunications Networks

China’s purported efforts, known as Salt Typhoon, to infiltrate American telecommunications companies and acquire data about U.S. calls were the subject of a classified briefing for all senators on Wednesday, according to Reuters. When the briefing was over, “US senators vow[ed] action.”

“A Senate Commerce subcommittee will hold a hearing on December 11 on Salt Typhoon and how security threats pose risks to our communications networks, and review best practices,'” according to Reuters.

Concern over the extent and magnitude of the alleged Chinese hacking of American telecommunications networks is growing, as is uncertainty about when businesses and the government will be able to reassure citizens about the situation.

According to reports, CISA’s Greene recommended “that Americans should use encrypted apps for all their communications” during Tuesday’s initial media briefing (1,2). This means that even while iMessages and Google Messages are completely secured on those platforms, you should avoid sending messages from your iPhone to Android.

“Our recommendation, which we have discussed internally, is not new here: encryption is your friend, whether you’re using text messaging or have the ability to use encrypted voice communication,” Greene continued.

Encrypting the data will prevent the adversary from intercepting it, even if they are able to do so.

The FBI, CISA, NSA, and other Five Eyes organisations together issued a notice on Tuesday regarding the ongoing telco network breaches.

One obvious deficiency is the absence of end-to-end encryption to safeguard cross-platform RCS, SMS’s replacement. Samsung recently issued a PR release celebrating the success of RCS, highlighting the fact that only texting between Android devices is safe.

End-to-end encryption for Android and iPhone

The fact that Google and Apple separately recommend end-to-end encryption for Android and iPhone users, but it is still absent from RCS and has no estimated time of arrival, is still a glaring irony.

Although there isn’t a specific date yet, Google, GSMA, and the mobile standard-setter have stated that encryption will eventually be included to RCS.

With the media picking up on the security vulnerability after Apple’s upgrade, that assurance appeared to be a reaction to the criticism. Apple, the company behind the more fully encrypted iPhone environment, has refrained from commenting.

These warnings have an ironic twist. The FBI has long complained that the same technology can impede their investigations into seized devices and online accounts belonging to criminal suspects, so PC Mag noted that “this push to use end-to-end encryption is ironic.”

In light of this, the FBI’s exact language is crucial, emphasising prudent encryption, which has otherwise received little attention in stories. In this context, “responsible” refers to granting access to user data, and possibly content, through legitimate requests.

FBI recommends Whatsapp

Although this might seem like a subtlety, it is not. Due to their inability to grant access to any content without an endpoint (device) compromise, which would allow access to the data at one end of the end-to-end encryption, this excludes many of the biggest and most well-known messaging systems, including WhatsApp and Signal.

However, until RCS adds its own full encryption across iPhones and Androids, I still advise using the fully encrypted WhatsApp over RCS for any cross-platform chat.

These security measures vanish as soon as you leave Google’s or Apple’s walled gardens. Now that there are so many excellent, secure platforms accessible, the risk is not worth it. Given the constantly changing world of cyber threats, complete security is more important than ever.

Other completely encrypted platforms exist as well; Signal is the greatest of them all, albeit its install base is far smaller. The fact that Facebook Messenger now completely encrypts messages further distinguishes ordinary SMS/RCS texting as an anomaly.

Given this FBI/CISA warning, Signal and WhatsApp should also be your first choices because they allow cross-platform, completely encrypted audio and video calls.

Ironically, iPhone owners will be able to switch from iMessage as the default messenger on their smartphones with Apple’s iOS 18.2, which is scheduled for release this month. It’s all about timing.

Continue Reading

Tech

The US FDA Cited Musk’s Neuralink Animal Laboratory For ‘Objectionable Conditions.’

Published

on

Neuralink
REUTERS/Dado Ruvic/Illustration/File Photo

(VOR News) – The Food and Drug Administration (FDA) discovered Neuralink “objectionable conditions or practices” at an animal testing facility under the brain technology business Neuralink.

One firm that manufactures brain technology is Neuralink. The Food and Drug Administration (FDA) reprimanded the corporation and recommended that it take corrective action in response to the laboratory deficiencies that were found.

According to the letter delivered to Congress on November 29th, Food and Drug Administration (FDA) inspectors discovered the issues at Neuralink’s animal testing facility in California in June 2023.

The letter that was sent contained this information. The Food and Drug Administration had given the company permission to perform a limited study of their brain implant in human volunteers a few weeks prior.

Neuralink’s incident happened a few weeks after FDA approval.

During the time that Oregon Democrat Earl Blumenauer raised the issue, the agency was responding to particular challenges raised by Blumenauer, who has long been concerned about animal welfare. The problems had been raised by Blumenauer.

The Food and Drug Administration (FDA) alerted Blumenauer to the issue that was found at Neuralink and told Blumenauer that the company had to do repair operations on its own initiative. The agency did not consider them substantial enough to justify pursuing any regulatory action against them. They were deemed insufficiently serious.

The following is a quote from a statement issued by the Food and Drug Administration (FDA) in response to enquiries from Reuters. You can view the statement here: “The FDA’s inspection did not find evidence of any violations that would undermine the device’s safety.”

This action was conducted in response to citations that were given to the organisation regarding the documentation of their animal study. It should be mentioned that no further details regarding the sites that are of concern are included in this report.

Reuters reported in February that during an inspection conducted in June 2023, FDA officials found problems with document keeping and quality standards at Neuralink’s laboratory in California. The facility’s record-keeping and quality control practices were the focus of the investigation.

The examination was conducted in the United States of America. In contrast, the Food and Drug Administration inspectors had not yet decided how much action should be taken.

The Food and Drug Administration (FDA) did not comment on the issue, nor did it address whether these patients were afflicted with the same medical conditions described in the letter.

Neuralink didn’t respond to our questions about FDA concerns.

The company has stated that it goes above and above the current regulations for the care of animals when it comes to the way it handles animals used for research.

According to reports, in 2022, Neuralink employees voiced their displeasure with the company’s hurried testing practices, which led to animal suffering and the deaths of pigs, monkeys, and other creatures that weren’t needed for the trials. These accusations were made public, according to Reuters.

Neuralink is one of several businesses striving to help those who are unable to interact with the outside world due to illnesses including paralysis, blindness, and other problems. One company that uses this technology to translate human ideas into digital communication is Neuralink, which uses a brain chip.

This business, which has already implanted its device in two patients in the US, intends to conduct a study project in Canada that is similar to the one it has already completed in the US.

Musk has been tasked by President-elect Donald Trump’s team with cutting back on government spending on agencies like the Food and Drug team (FDA). The government of President-elect Trump appointed Musk to this role.

His role as the director of the newly created “Department of Government Efficiency” may help shield his own companies from additional regulation or enforcement and increase the government’s backing for them, according to Reuters. The news agency provided this information. A piece that was published precisely one month ago revealed this information.

SOURCE: USN

SEE ALSO:

Canadian Media Firms Are Suing OpenAI in a Potential Billion-Dollar Dispute.

Increasing its Stake in OpenAI by $1.5 Billion is a Possibility for SoftBank.

Continue Reading

Trending