Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is in response to the request for continued examination filed on
February 20, 2026. Claims 1-20 are pending. Claims 1-20 represent EARLY DETECTION OF PROMPT INJECTION ATTACKS USING SEMANTIC ANALYSIS.
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
2. Claims 1, 11 and 20 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the enablement requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to enable one skilled in the art to which it pertains, or with which it is most nearly connected, to make and/or use the invention. In page 1, lines 4-7 of claim 1, and similarly with claims 11 and 20, Applicant is claiming subject matter that is not describe in the discloses. Appropriate correction is required.
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
3. Claims 1, 11 and 20 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as failing to set forth the subject matter which the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the applicant regards as the invention. Evidence that claims 1, 11 and 20 fail to correspond in scope with that which the inventor or a joint inventor, or for pre-AIA applications the applicant regards as the invention can be found in the reply filed February 20, 2026. In that paper, the inventor or a joint inventor, or for pre-AIA applications the applicant has stated “intra-prompt” and wherein the variation is computed by comparing vector representations of words within the prompt, as obtained, to one another,” is not disclosed in the specification and this statement indicates that the invention is different from what is defined in the claim(s) because Examiner did not find such teaching in the specification.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
4. Claims 1-20 are rejected under 35 U.S.C. 103 as being unpatentable over Gardner U.S. 20240296279 and further in view of Maddux U.S. 20250173438.
Gardner teaches the invention substantially including confidence enhancement for responses by document-based large language models (see abstract).
As to claim 1, Gardner teaches a method comprising:
obtaining, by a device, a prompt for input to a language model (abstract, Garner disclose a prompt is generated and the prompt is used by a language model);
identifying, by the device, a plurality of topics present in the prompt (paragraph 99, Gardner prompt 445 including the dialog context and the set of similar example pairs (I.e. “plurality of topics present in the prompt”); and
preventing, by the device, the prompt from being processed by the language model (paragraph 99, Gardner discloses if the current dialogue context is a malicious or adverse input, a second AI Model 450, to which the second prompt 445 is to be provided, is more likely to produce a proper or mitigating response rather than an improper response…).
Gardner teaches substantial features of the claimed invention but fails to explicitly teach determining, by the device, that the prompt is malicious based on intra-prompt variation among the plurality of topics, wherein the variation is computed by comparing vector representations of words within the prompt, as obtained, to one another; and.
However, Maddux teaches system and method for detecting prompt injection attacks to large language models. Maddux teaches determining, by the device, that the prompt is malicious based on intra-prompt variation among the plurality of topics, wherein the variation is computed by comparing vector representations of words within the prompt, as obtained, to one another (paragraph 158, Maddux discloses Neural network 862 may be configured to identify or detect prompt injection attacks (i.e. plurality of topics present within the prompt) based on the attention scores. Neural network 862 may be substantially similar to neural network 262. For example, neural network 862 may be trained to analyze the attention scores to determine whether the prompt is malicious. In some embodiments, such analysis may involve determining whether the prompt is more highly correlated with the negative contexts than one or more of the positive contexts and itself).
It would have been obvious to one of ordinary skill in the art before the effective filing date to modify the method of Gardner to include the teaching of “determining, by the device, that the prompt is malicious based on a variation among the plurality of topics present within the prompt” of Maddux with the motivation being determining that the prompt is likely a malicious prompt, and the computing system initiates a remedial action in response (see abstract).
As to claim 2, Gardner and Maddux teach the method as in claim 1, wherein the language model is a large language model trained to perform a plurality of tasks (paragraph 4).
As to claim 3, Gardner and Maddux teach the method as in claim 1, wherein determining that the prompt is malicious based on a variation in the plurality of topics comprises: converting words present in the prompt into vector representations; and computing distance metrics between the vector representations (paragraph 94).
As to claim 4, Gardner and Maddux teach the method as in claim 1, further comprising: providing, by the device and to a user interface, an indication that the prompt is a suspected prompt injection attack (paragraph 7).
As to claim 5, Gardner and Maddux teach the method as in claim 1, further comprising: determining, by the device, a measure of sentence incoherence of the prompt by evaluating any noun-adjective or verb-adverb word pairs in the prompt, wherein the device determines that the prompt is malicious based in part on the measure of sentence incoherence (paragraph 7).
As to claim 6, Gardner and Maddux teach the method as in claim 5, further comprising: computing the measure of sentence incoherence in part by comparing any noun- adjective or verb-adverb word pairs in the prompt to noun-adjective or verb-adverb word pairs in a baseline corpus (paragraph 96).
As to claim 7, Gardner and Maddux teach the method as in claim 5, further comprising: providing, by the device, an indication that the prompt was prevented from being processed by the language model because of the measure of sentence incoherence (paragraph 81).
As to claim 8, Gardner and Maddux teach the method as in claim 1, wherein the language model is configured to interact with a network controller (paragraph 130).
As to claim 9, Gardner and Maddux teach the method as in claim 1, wherein obtaining the prompt for input to the language model comprises: intercepting, by the device, the prompt prior to input to the language model (paragraph 49).
As to claim 10, Gardner and Maddux teach the method as in claim 1, wherein the device determines that the prompt is malicious without analyzing the prompt using a machine learning-based model (paragraph 79).
Claims 11-20 did not teach anything different from above rejected claims 1-10, therefore are rejected similarly.
Conclusion
5. Any inquiry concerning this communication or earlier communications from the examiner should be directed to EL HADJI SALL whose telephone number is (571)272-4010. The examiner can normally be reached Monday-Friday 8:00-8:30 (flexible).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Ario Etienne can be reached at 5712724001. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/EL HADJI M SALL/Primary Examiner, Art Unit 2457